diff --git a/apps/sim/lib/compare/data/competitors/crewai.ts b/apps/sim/lib/compare/data/competitors/crewai.ts
index ea230014aaa..6924cb95276 100644
--- a/apps/sim/lib/compare/data/competitors/crewai.ts
+++ b/apps/sim/lib/compare/data/competitors/crewai.ts
@@ -51,10 +51,11 @@ export const crewaiProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Native Agent2Agent (A2A) protocol support as a first-class primitive',
+      title: 'Acts as both an A2A client and an A2A server',
       description:
-        'CrewAI treats the open Agent2Agent (A2A) protocol as a first-class delegation primitive: agents can be configured with an A2AClientConfig to delegate tasks to and request information from remote A2A-compliant agents (with Bearer, OAuth2, API key, or HTTP auth), and/or an A2AServerConfig to expose a CrewAI agent as an A2A-compliant server other frameworks can call, via the optional crewai[a2a] extra.',
-      shortDescription: 'Delegates to and serves as remote agents via the open A2A protocol.',
+        'CrewAI treats the open Agent2Agent (A2A) protocol as a first-class delegation primitive: agents can be configured with an A2AClientConfig to delegate tasks to and request information from remote A2A-compliant agents (with Bearer, OAuth2, API key, or HTTP auth), and/or an A2AServerConfig to expose a CrewAI agent as an A2A-compliant server other frameworks can call, via the optional crewai[a2a] extra. Sim ships a dedicated A2A block that calls, tracks, and discovers external A2A-compliant agents, but does not document a way to expose a Sim workflow as an A2A server of its own.',
+      shortDescription:
+        "Delegates to remote A2A agents and can expose a crew as an A2A server; Sim's A2A block only calls out to external agents.",
       source: {
         url: 'https://docs.crewai.com/en/learn/a2a-agent-delegation',
         label: 'Agent-to-Agent (A2A) Protocol - CrewAI Docs',
@@ -64,9 +65,9 @@ export const crewaiProfile: CompetitorProfile = {
     {
       title: 'CrewAI AMP: natural-language visual Studio on top of the code framework',
       description:
-        'CrewAI AMP (the commercial Agent Management Platform) adds Crew Studio, a chat-and-canvas interface where a builder describes an automation in natural language and the AI generates agents, tasks, and tools as an editable drag-and-drop workflow, exportable to Python code. This gives the code-first framework an optional visual entry point for non-developers.',
+        'CrewAI AMP (the commercial Agent Management Platform) adds Crew Studio, a chat-and-canvas interface where a builder describes an automation in natural language and the AI generates agents, tasks, and tools as an editable drag-and-drop workflow, exportable to Python code. This gives the code-first framework an optional visual entry point for non-developers. Sim ships an equivalent natural-language builder (Chat and in-editor Copilot) as a core, free part of the product, not a separate paid add-on layered on top of a code-only open-source base.',
       shortDescription:
-        'Natural-language chat generates an editable visual workflow, exportable to code.',
+        "Natural-language chat generates an editable visual workflow, exportable to code, as a paid AMP add-on; Sim's Chat and Copilot ship the same capability free.",
       source: {
         url: 'https://docs.crewai.com/en/enterprise/features/crew-studio',
         label: 'Crew Studio - CrewAI Docs',
@@ -110,17 +111,6 @@ export const crewaiProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
-    {
-      title: 'Native vector store support limited to two backends',
-      description:
-        "CrewAI's built-in RAG/knowledge system ships native support for only ChromaDB (the default) and Qdrant as vector store backends. Broader coverage (Pinecone, PGVector, Supabase, etc.) requires custom integration work, not a documented first-party connector.",
-      shortDescription: 'Native knowledge/RAG vector stores are limited to ChromaDB and Qdrant.',
-      source: {
-        url: 'https://docs.crewai.com/en/concepts/knowledge',
-        label: 'Knowledge - CrewAI Docs',
-        asOf: '2026-07-02',
-      },
-    },
     {
       title: 'Requires Python fluency; no low-code entry point in the core product',
       description:
diff --git a/apps/sim/lib/compare/data/competitors/dust.ts b/apps/sim/lib/compare/data/competitors/dust.ts
index a93120f9540..3e1bb4ed63d 100644
--- a/apps/sim/lib/compare/data/competitors/dust.ts
+++ b/apps/sim/lib/compare/data/competitors/dust.ts
@@ -17,10 +17,11 @@ export const dustProfile: CompetitorProfile = {
     'Dust is an enterprise AI agent platform where teams build no-code agents connected to company data and tools in a shared, multiplayer workspace, then deploy them to chat, Slack, and other surfaces.',
   standoutFeatures: [
     {
-      title: 'Purely no-code, form-based builder for non-technical teams',
+      title:
+        'Zero visual/flow layer by design, building only through forms, text, and conversation',
       description:
-        "Dust's Agent Builder is entirely form and text based, name, description, instructions, model, tools, knowledge, guided by a conversational 'Sidekick' assistant, with no visual canvas at all (its earlier block-based 'Dust Apps' product is deprecated). Agents deploy natively into a shared, multiplayer workspace and out to Slack, Teams, and other chat surfaces. A team that wants business users assembling agents from plain-language instructions and templates, with no drag-and-drop layer to learn, gets that directly. Teams that do want infrastructure-as-code can also define Skills and agent configurations as files in a Git repository and sync them via an official GitHub Action, with the same PR review and rollback workflow as application code.",
-      shortDescription: 'No-code, form-based builder for business teams, no visual canvas at all.',
+        "Dust's Agent Builder is entirely form and text based, name, description, instructions, model, tools, knowledge, guided by a conversational 'Sidekick' assistant, with no visual canvas at all (its earlier block-based 'Dust Apps' product is deprecated). Agents deploy natively into a shared, multiplayer workspace and out to Slack, Teams, and other chat surfaces. A team that wants agents assembled purely from plain-language instructions and templates, with no drag-and-drop layer to learn or maintain, gets that directly. Teams that do want infrastructure-as-code can also define Skills and agent configurations as files in a Git repository and sync them via an official GitHub Action, with the same PR review and rollback workflow as application code.",
+      shortDescription: 'No visual/flow canvas at all, only forms, text, and conversation.',
       source: {
         url: 'https://docs.dust.tt/changelog/gitops-sync-for-skills-agent-configurations-with-github-action',
         label: 'GitOps sync for Skills & Agent configurations | Dust changelog',
@@ -28,11 +29,10 @@ export const dustProfile: CompetitorProfile = {
       },
     },
     {
-      title: "'Skills' as reusable, shared agent instruction/tool packages",
+      title: "'Skills' can attach to many agents at once, with one edit propagating to all of them",
       description:
-        "Skills are named, reusable packages of instructions, knowledge, and tools that can be attached to multiple agents at once. Updating a Skill's instructions automatically propagates the improvement to every agent using it, rather than requiring each agent to be edited individually.",
-      shortDescription:
-        'Reusable instruction/tool packages that update every agent using them at once.',
+        "A single Skill can be attached to multiple agents simultaneously, and updating its instructions once automatically propagates that change to every agent using it, rather than requiring each agent's copy to be edited individually.",
+      shortDescription: 'One Skill edit auto-propagates to every agent it is attached to.',
       source: {
         url: 'https://docs.dust.tt/docs/skills',
         label: 'Skills | Dust Docs',
@@ -51,10 +51,11 @@ export const dustProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Dual-role MCP: consumes external servers and exposes Dust as one',
+      title: "Client-side MCP tools that execute locally in the end user's own environment",
       description:
-        "Dust agents can call tools from external MCP servers, remote or client-side (client-side tools execute in the user's own environment for sensitive operations). Dust can also be exposed as an MCP server, so external MCP-compatible clients (e.g. Claude Desktop, Cursor) can call Dust agents and data as tools.",
-      shortDescription: 'Both calls external MCP tools and can be called as an MCP server itself.',
+        "Beyond remote MCP servers, Dust supports client-side MCP servers whose tools run directly in the end user's local environment rather than on Dust's own infrastructure, for sensitive operations that shouldn't leave the user's machine. Dust can also be exposed as an MCP server itself, so external MCP-compatible clients (e.g. Claude Desktop, Cursor) can call Dust agents and data as tools.",
+      shortDescription:
+        "MCP tools that run locally in the user's own environment for sensitive operations.",
       source: {
         url: 'https://docs.dust.tt/docs/client-side-mcp-server',
         label: 'Client Side MCP Server (Preview) | Dust Docs',
@@ -87,10 +88,12 @@ export const dustProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'No dedicated pre-deployment evaluation/dataset-testing framework',
+      title:
+        'No dedicated pre-deployment evaluation/dataset-testing framework, a gap shared with most agent builders',
       description:
-        "Dust says it is 'not a pre-deployment evaluation platform': dataset-based regression testing belongs in CI/CD pipelines and specialized testing tools, and Dust builds observability signals into the agent-builder workflow instead of a formal eval-suite feature.",
-      shortDescription: 'Dust says it is not a pre-deployment evaluation platform.',
+        "Dust explicitly says it is 'not a pre-deployment evaluation platform': dataset-based regression testing belongs in CI/CD pipelines and specialized testing tools, and Dust builds observability signals into the agent-builder workflow instead of a formal eval-suite feature. This is a gap most agent builders share, including Sim, whose own Evaluator and Guardrails blocks are per-call scoring/validation primitives rather than a batch golden-dataset eval-suite runner.",
+      shortDescription:
+        'Dust says it is not a pre-deployment eval platform, a gap shared with most agent builders.',
       source: {
         url: 'https://dust.tt/blog/evaluation-to-maintenance',
         label: 'From Evaluation to Maintenance | Dust Blog',
diff --git a/apps/sim/lib/compare/data/competitors/flowise.ts b/apps/sim/lib/compare/data/competitors/flowise.ts
index e3ab62ae007..3ed440bee2d 100644
--- a/apps/sim/lib/compare/data/competitors/flowise.ts
+++ b/apps/sim/lib/compare/data/competitors/flowise.ts
@@ -29,23 +29,23 @@ export const flowiseProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Agentflow V2 with built-in human-in-the-loop and evaluation',
+      title: 'Built-in dataset-based batch evaluation',
       description:
-        'Agentflow V2 supports loops, conditional branching, and a dedicated Human Input node that pauses execution for approve/reject feedback before sensitive tool calls (bookings, sends, orders) proceed. Flowise also ships a built-in Evaluations feature that runs chatflows/agentflows against a dataset, scoring outputs with string, numeric, or LLM-as-judge evaluators and reporting pass/fail rate, average tokens, and latency.',
+        "Flowise ships a built-in Evaluations feature that runs chatflows/agentflows against a saved dataset in one batch, scoring outputs with string, numeric, or LLM-as-judge evaluators and reporting pass/fail rate, average tokens, and latency across the whole run. Sim's own Evaluator block scores individual calls against user-defined metrics, but has no equivalent golden-dataset batch runner. (Flowise's Agentflow V2 also has a Human Input node for pausing on approve/reject feedback, comparable to Sim's own human-in-the-loop approval block.)",
       shortDescription:
-        'Native human-approval node plus built-in dataset-based LLM-judge evaluation reporting.',
+        'Built-in dataset-based batch evaluation with LLM-judge scoring and pass/fail reporting.',
       source: {
-        url: 'https://docs.flowiseai.com/tutorials/human-in-the-loop',
-        label: 'Flowise Docs: Human In The Loop',
+        url: 'https://docs.flowiseai.com/using-flowise/evaluations',
+        label: 'Flowise Docs: Evaluations',
         asOf: '2026-07-02',
       },
     },
     {
-      title: 'Large open-source project with Apache 2.0 core',
+      title: 'Larger existing open-source community, on the same Apache 2.0 license as Sim',
       description:
-        "Flowise's Community Edition is Apache License 2.0, its GitHub repo has roughly 54,000 stars, and it has an active Discord community with full self-hosting support via Docker.",
+        'Both Flowise and Sim are Apache License 2.0 and self-hostable, so the license itself is not a differentiator. Where Flowise stands out is community scale: its GitHub repo has roughly 54,000 stars and an active Discord community built up since 2023.',
       shortDescription:
-        'Apache 2.0 licensed, ~54k GitHub stars, actively maintained open-source project.',
+        'Same Apache 2.0 license as Sim, but a larger existing community: ~54k GitHub stars, active Discord.',
       source: {
         url: 'https://github.com/FlowiseAI/Flowise',
         label: 'GitHub: FlowiseAI/Flowise',
diff --git a/apps/sim/lib/compare/data/competitors/gumloop.ts b/apps/sim/lib/compare/data/competitors/gumloop.ts
index 2b444e5810f..edbebc9f1ee 100644
--- a/apps/sim/lib/compare/data/competitors/gumloop.ts
+++ b/apps/sim/lib/compare/data/competitors/gumloop.ts
@@ -37,11 +37,10 @@ export const gumloopProfile: CompetitorProfile = {
       },
     },
     {
-      title: "Natural-language flow building via 'Gen' copilot",
+      title: 'Gen copilot debugs existing flows node-by-node on canvas',
       description:
-        'An AI assistant named Gen can plan, build, modify, and debug workflows from plain-English descriptions directly on the canvas, lowering the barrier for non-technical users.',
-      shortDescription:
-        'An AI copilot builds, edits, and debugs workflows from plain-English prompts.',
+        'Beyond building new flows from a prompt, Gen can step into an already-built workflow and debug it node-by-node directly on the canvas, working from a plain-English description of what is going wrong.',
+      shortDescription: 'Gen debugs an existing workflow node-by-node directly on the canvas.',
       source: {
         url: 'https://www.gumloop.com/blog/agentic-ai-tools',
         label: 'Gumloop blog: agentic AI tools',
@@ -49,11 +48,10 @@ export const gumloopProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Plain-English guardrail policies with human-in-the-loop approval',
+      title: 'Plain-English, org-wide guardrail policy engine',
       description:
-        'Organizations can define app/tool usage policies in plain English at org, team, or agent level; violating actions can be blocked or tagged and logged, and sensitive actions can pause mid-task for a human approval card.',
-      shortDescription:
-        'Plain-English usage policies plus mid-task human approval for sensitive actions.',
+        'Organizations can define app/tool usage policies in plain English at org, team, or agent level; violating actions can be blocked or tagged and logged.',
+      shortDescription: 'Plain-English usage policies enforced at org, team, or agent level.',
       source: {
         url: 'https://www.gumloop.com/solutions/security',
         label: 'Gumloop Security & Trust',
@@ -88,7 +86,7 @@ export const gumloopProfile: CompetitorProfile = {
     {
       title: 'No public self-hosting of the core platform',
       description:
-        "Gumloop is only available as managed SaaS or an enterprise-managed VPC deployment operated by Gumloop inside a customer's cloud project. There is no downloadable, self-managed install of the Gumloop application itself; a separate community project, guMCP, is open source but is not the platform.",
+        "Gumloop is only available as managed SaaS or an enterprise-managed VPC deployment operated by Gumloop inside a customer's cloud project. There is no downloadable, self-managed install of the Gumloop application itself; Gumloop's own guMCP_template repo is a self-hosted MCP-server starter, not an install of the platform.",
       shortDescription: 'No downloadable self-hosted install. Only managed SaaS or enterprise VPC.',
       source: {
         url: 'https://www.gumloop.com/solutions/security',
@@ -163,7 +161,7 @@ export const gumloopProfile: CompetitorProfile = {
         value:
           'No public self-host option for the core Gumloop app; enterprise customers can get a managed Virtual Private Cloud (VPC) deployment into their own cloud (e.g. GCP) instead of full self-hosting',
         detail:
-          "Gumloop deploys and operates the platform inside the customer's cloud project rather than offering a downloadable, self-managed open-source install. A separate community open-source project, guMCP, provides self-hostable MCP servers but is not the Gumloop app itself.",
+          "Gumloop deploys and operates the platform inside the customer's cloud project rather than offering a downloadable, self-managed open-source install. Gumloop's own guMCP_template repo is an open-source starter for self-hosted MCP servers, but it is not an install of the Gumloop app itself.",
         shortValue: 'No self-host; VPC deployment only',
         confidence: 'estimated',
         sources: [
@@ -173,8 +171,8 @@ export const gumloopProfile: CompetitorProfile = {
             asOf: '2026-07-02',
           },
           {
-            url: 'https://github.com/agoddijn-fern/guMCP-self-hosted',
-            label: 'guMCP self-hosted (community project)',
+            url: 'https://github.com/gumloop/guMCP_template',
+            label: "guMCP_template (Gumloop's self-hosted MCP starter repo)",
             asOf: '2026-07-02',
           },
         ],
@@ -210,7 +208,7 @@ export const gumloopProfile: CompetitorProfile = {
       license: {
         value: 'Proprietary',
         detail:
-          'The core Gumloop application has no open-source license; it is a closed, hosted commercial SaaS product. A separate community MCP-server project, guMCP, is open source but is not the Gumloop platform.',
+          "The core Gumloop application has no open-source license; it is a closed, hosted commercial SaaS product. Gumloop's own guMCP_template repo is an open-source MCP-server starter, but it is not the Gumloop platform.",
         shortValue: 'Proprietary',
         confidence: 'estimated',
         sources: [
@@ -722,6 +720,8 @@ export const gumloopProfile: CompetitorProfile = {
       entryPaidPlan: {
         value:
           'Pro plan at $37/month for 20k+ credits. Includes unlimited seats/teams, 5 concurrent runs, 25 concurrent agent interactions, agent reflections, unified billing, and 1 hosted MCP server instance',
+        detail:
+          "Gumloop's pricing page lists 'MCP Server Hosting (1)' under the Pro plan without clarifying its scope: it is not stated whether this cap limits access to the 100+ pre-built, zero-setup MCP servers described on gumloop.com/mcp, or only applies to a separate custom MCP server that Gumloop hosts on a customer's behalf. That distinction is not resolved anywhere on Gumloop's own pricing or MCP pages.",
         shortValue: '$37/month Pro plan, 20k+ credits',
         confidence: 'verified',
         sources: [
@@ -1117,11 +1117,11 @@ export const gumloopProfile: CompetitorProfile = {
       },
       companyMaturity: {
         value:
-          "Founded in Vancouver in April 2023 (originally as 'AgentHub') by Max Brodeur-Urbas and Rahul Behal. Raised a $3.1M seed (July 2024), a $17M Series A in January 2025 (led by Nexus Venture Partners), and a $50M Series B in March 2026 (led by Benchmark). About $70M raised in total across 3 rounds. Y Combinator alum with roughly 37 employees as of mid-2026.",
+          "Founded in Vancouver in April 2023 (originally as 'AgentHub') by Max Brodeur-Urbas and Rahul Behal. Raised a $3.1M seed (July 2024) and a $17M Series A in January 2025 (led by Nexus Venture Partners), both independently corroborated; a self-reported $50M Series B in March 2026 (led by Benchmark) would bring the total to about $70M across 3 rounds. Y Combinator alum with roughly 37 employees as of mid-2026.",
         detail:
-          'Gumloop started as a side project in a Vancouver bedroom in April 2023, founded by Max Brodeur-Urbas and Rahul Behal under the name AgentHub before rebranding to Gumloop. It raised a $3.1M seed round in July 2024, a $17M Series A in January 2025 led by Nexus Venture Partners (with First Round Capital, Y Combinator, and angel investors), and a $50M Series B in March 2026 led by Benchmark (with Nexus Venture Partners, First Round Capital, Y Combinator, Box Group, The Cannon Project, and Shopify Ventures). Total raised is about $70M across 3 rounds. Y Combinator lists a team size of 37.',
-        shortValue: 'Founded 2023, ~$70M raised, Series B in 2026',
-        confidence: 'verified',
+          "Gumloop started as a side project in a Vancouver bedroom in April 2023, founded by Max Brodeur-Urbas and Rahul Behal under the name AgentHub before rebranding to Gumloop. It raised a $3.1M seed round in July 2024 and a $17M Series A in January 2025 led by Nexus Venture Partners (with First Round Capital, Y Combinator, and angel investors), both independently corroborated by TechCrunch. The $50M Series B in March 2026 led by Benchmark (with Nexus Venture Partners, First Round Capital, Y Combinator, Box Group, The Cannon Project, and Shopify Ventures) is self-reported on Gumloop's own blog only, with no independent press or funding-tracker corroboration found. Total raised is about $70M across 3 rounds. Y Combinator lists a team size of 37.",
+        shortValue: 'Founded 2023, ~$70M raised, Series B in 2026 (self-reported)',
+        confidence: 'estimated',
         sources: [
           {
             url: 'https://www.gumloop.com/blog/gumloops-17m-series-a',
diff --git a/apps/sim/lib/compare/data/competitors/langchain.ts b/apps/sim/lib/compare/data/competitors/langchain.ts
index 2bdd4955168..0c58d2b12be 100644
--- a/apps/sim/lib/compare/data/competitors/langchain.ts
+++ b/apps/sim/lib/compare/data/competitors/langchain.ts
@@ -32,7 +32,7 @@ export const langchainProfile: CompetitorProfile = {
     {
       title: 'Dynamic parallel fan-out via the Send API',
       description:
-        'A routing function can return a list of Send objects instead of a single next-node key, letting LangGraph spawn a runtime-determined number of parallel branches (e.g. one worker per item in a list of unknown length) that merge back through a state reducer. This is a native map-reduce pattern, not a fixed number of parallel branches wired at build time.',
+        'A routing function can return a list of Send objects instead of a single next-node key, letting LangGraph spawn a runtime-determined number of parallel branches (e.g. one worker per item in a list of unknown length) that merge back through a state reducer. Because the merge step is arbitrary code, a developer can implement any custom aggregation logic, not just a fixed join behavior.',
       shortDescription:
         'Send API spawns a runtime-determined number of parallel branches that merge via a reducer.',
       source: {
@@ -53,18 +53,6 @@ export const langchainProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
-    {
-      title: 'SKILL.md-based reusable agent skills via Deep Agents SkillsMiddleware',
-      description:
-        "The Deep Agents harness (LangChain's batteries-included agent framework) ships a SkillsMiddleware that loads named SKILL.md files from a directory and injects them into the system prompt using progressive disclosure (metadata surfaced first, full instructions pulled in on demand), letting a team define a workflow once and reuse it as a named capability across multiple agents.",
-      shortDescription:
-        'SkillsMiddleware loads named SKILL.md files and injects them via progressive disclosure.',
-      source: {
-        url: 'https://reference.langchain.com/python/deepagents/middleware/skills',
-        label: 'skills | deepagents | LangChain Reference',
-        asOf: '2026-07-02',
-      },
-    },
     {
       title: 'Native A2A protocol support as both server and client',
       description:
@@ -114,18 +102,6 @@ export const langchainProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
-    {
-      title: 'Durability is checkpoint persistence, not automatic failure detection',
-      description:
-        "LangGraph's checkpointer saves state after every node, but nothing automatically detects a crashed process; it only lets a resumed process recover from the last saved state. An operator (or external process supervisor) still has to notice the failure and trigger the resume.",
-      shortDescription:
-        'Checkpointer saves state on failure, but nothing automatically detects a crashed process.',
-      source: {
-        url: 'https://www.diagrid.io/blog/checkpoints-are-not-durable-execution-why-langgraph-crewai-google-adk-and-others-fall-short-for-production-agent-workflows',
-        label: "Why Checkpoints Aren't Durable Execution: LangGraph",
-        asOf: '2026-07-02',
-      },
-    },
     {
       title: 'No dedicated native image/video/audio generation capability',
       description:
@@ -564,7 +540,7 @@ export const langchainProfile: CompetitorProfile = {
       },
       a2aProtocol: {
         value:
-          "Yes: LangChain shipped native A2A (Agent2Agent) support via langchain-adk (March 2026), letting any LangChain agent expose itself as an A2A server and call other A2A-compliant agents regardless of the framework that built them, with Agent Cards auto-generated from the agent's name/description/tool list. The local LangGraph dev server exposes A2A endpoints at /a2a/{assistant_id}.",
+          "Yes: the local LangGraph dev server and the LangSmith Deployment Agent Server both expose native A2A (Agent2Agent) endpoints at /a2a/{assistant_id}, letting any LangChain/LangGraph agent expose itself as an A2A server and call other A2A-compliant agents regardless of the framework that built them, with Agent Cards auto-generated from the agent's name/description/tool list.",
         detail:
           "The LangSmith Deployment A2A endpoint maps the protocol's contextId to a LangGraph thread_id automatically, so A2A conversations get the same tracing/observability as native LangGraph runs.",
         shortValue: 'Yes, native A2A server/client support with auto-generated Agent Cards',
diff --git a/apps/sim/lib/compare/data/competitors/langflow.ts b/apps/sim/lib/compare/data/competitors/langflow.ts
index 80a89675709..93e28c1aee0 100644
--- a/apps/sim/lib/compare/data/competitors/langflow.ts
+++ b/apps/sim/lib/compare/data/competitors/langflow.ts
@@ -29,10 +29,11 @@ export const langflowProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Dual-direction MCP support',
+      title: 'Zero-step MCP server exposure',
       description:
-        'Langflow can act as an MCP client, connecting to external MCP servers as tool sources. It also automatically exposes every flow with a Chat Output as its own MCP server, so any flow becomes a callable tool for outside MCP clients.',
-      shortDescription: 'Both consumes external MCP servers and publishes flows as MCP servers.',
+        'Every Langflow project is automatically registered as an MCP server the moment it is created, exposing any flow with a Chat Output as a callable MCP tool with no separate publish or deploy step. Sim also supports both MCP client and server roles, but publishing a workflow as an MCP tool requires an explicit deploy step.',
+      shortDescription:
+        'Every project auto-registers as an MCP server on creation, no deploy step.',
       source: {
         url: 'https://docs.langflow.org/mcp-server',
         label: 'Langflow Docs: Use Langflow as an MCP server',
diff --git a/apps/sim/lib/compare/data/competitors/make.ts b/apps/sim/lib/compare/data/competitors/make.ts
index a66783f4d35..ee39cd274d7 100644
--- a/apps/sim/lib/compare/data/competitors/make.ts
+++ b/apps/sim/lib/compare/data/competitors/make.ts
@@ -31,10 +31,10 @@ export const makeProfile: CompetitorProfile = {
     'Make (make.com) is a closed-source, cloud-only visual workflow-automation platform where users connect app "modules" on a canvas into scenarios. It now also offers AI Agent blocks, an MCP server, and a JS/Python code step, billed on a per-module-execution credit model.',
   standoutFeatures: [
     {
-      title: '8,000+ template gallery available on the free tier',
+      title: '3,000+ integrations and an 8,000+ template gallery',
       description:
-        "Make's public template gallery hosts over 8,000 pre-built scenarios spanning thousands of use cases, browsable and importable free on every plan including Free, with users paying only for the credits consumed when the imported scenario runs.",
-      shortDescription: '8,000+ importable scenario templates, free on every plan including Free.',
+        'Make lists 3,000+ integration apps and a public gallery of over 8,000 pre-built, importable scenario templates, free to browse on every plan including Free.',
+      shortDescription: '3,000+ integrations and an 8,000+ template gallery, free on every plan.',
       source: {
         url: 'https://www.make.com/en/templates',
         label: 'Make Templates gallery',
@@ -53,35 +53,13 @@ export const makeProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Native MCP Server',
+      title: 'MCP Toolboxes for team-level shared servers',
       description:
-        'Make ships a first-party, cloud-hosted Model Context Protocol server that exposes any scenario as a callable tool to external AI agents/clients (Claude, Cursor, etc.) via a generated token and URL, with no local infrastructure to manage.',
-      shortDescription: 'Cloud-hosted MCP server exposes scenarios as tools with zero setup.',
+        'Beyond exposing a single scenario as an MCP tool, Make offers MCP Toolboxes: team-level dedicated MCP servers that bundle a curated subset of multiple scenarios behind one shared endpoint for external AI clients to call.',
+      shortDescription: 'Team-level MCP servers bundling multiple scenarios as tools.',
       source: {
-        url: 'https://www.make.com/en/blog/model-context-protocol-mcp-server',
-        label: 'Make blog: What is MCP Server?',
-        asOf: '2026-07-02',
-      },
-    },
-    {
-      title: 'Built-in Knowledge/RAG store for agents',
-      description:
-        "Users upload files directly to an agent's 'Knowledge,' which Make automatically chunks, embeds, and stores in a RAG vector database so the agent retrieves only relevant chunks at inference time, reducing token usage.",
-      shortDescription: 'Uploaded files are auto-chunked and embedded into a RAG store for agents.',
-      source: {
-        url: 'https://help.make.com/knowledge',
-        label: 'Make Knowledge help doc',
-        asOf: '2026-07-02',
-      },
-    },
-    {
-      title: 'In-scenario JS/Python code step',
-      description:
-        "The native 'Make Code' module lets users execute arbitrary JavaScript (Node.js) or Python directly inside a scenario step with no external server, for custom transforms and logic beyond the visual modules.",
-      shortDescription: 'Run JS or Python inline as a scenario step, no external server needed.',
-      source: {
-        url: 'https://www.make.com/en/blog/make-code-app',
-        label: 'Make blog: Make Code App',
+        url: 'https://help.make.com/mcp-toolboxes',
+        label: 'MCP toolboxes - Make Help Center',
         asOf: '2026-07-02',
       },
     },
@@ -121,10 +99,10 @@ export const makeProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Audit logs and granular RBAC gated to higher tiers',
+      title: 'Granular RBAC gated to the Teams plan and above',
       description:
-        'Audit logs and full team/role-based permission management are only available starting on the Teams plan ($38/mo) and Enterprise, with default log retention of just 30 days even on paid plans; lower tiers (Free, Core, Pro) lack these governance controls.',
-      shortDescription: 'Audit logs and role-based permissions require the Teams plan or above.',
+        "Full team/role-based permission management is only available starting on the Teams plan ($38/mo) and Enterprise; lower tiers (Free, Core, Pro) get unlimited users but no role-based access controls, unlike Sim, which ships admin/write/read roles on every tier. Audit logs are also gated to Teams/Enterprise, with default retention of just 30 days on paid plans, though this specific gating pattern isn't unique to Make: Sim's own dedicated audit-log API is likewise Enterprise-only.",
+      shortDescription: 'RBAC needs the Teams plan or above; audit logs are similarly gated.',
       source: {
         url: 'https://www.make.com/en/pricing',
         label: 'Make Pricing page',
diff --git a/apps/sim/lib/compare/data/competitors/microsoft-copilot.ts b/apps/sim/lib/compare/data/competitors/microsoft-copilot.ts
index cd799eae375..06d02a37abe 100644
--- a/apps/sim/lib/compare/data/competitors/microsoft-copilot.ts
+++ b/apps/sim/lib/compare/data/competitors/microsoft-copilot.ts
@@ -42,22 +42,11 @@ export const microsoftCopilotProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Reusable, portable Agent Skills',
+      title: 'Automatic fallback to a default model if a selected model is disabled',
       description:
-        'A Skill is a named capability defined once as a SKILL.md file (YAML front matter plus Markdown instructions, optionally bundled with scripts, templates, or reference documents into a ZIP package). Skills are authored in Copilot Studio or a text editor, attached to multiple agents, and exported to share with others, unlike a one-off system prompt tied to a single agent.',
-      shortDescription: 'Named, Markdown-defined capabilities reusable across multiple agents.',
-      source: {
-        url: 'https://learn.microsoft.com/en-us/microsoft-copilot-studio/agents-experience/skills-overview',
-        label: 'Skills overview for agents (preview) - Microsoft Copilot Studio | Microsoft Learn',
-        asOf: '2026-07-02',
-      },
-    },
-    {
-      title: 'Multi-model support including Anthropic Claude and bring-your-own-model',
-      description:
-        'Agents can use OpenAI GPT models, Anthropic Claude Sonnet 4 and Opus 4.1, any model in the Azure AI Model Catalog, or a bring-your-own-model connection to an Azure AI Foundry deployment (endpoint URI, deployment name, and API key) for individual prompts. Admins enable or restrict non-default models tenant-wide, and the agent falls back to the default OpenAI model automatically if a selected model is disabled.',
+        "If an admin disables a non-default model tenant-wide, Copilot Studio automatically falls back to the default OpenAI model rather than failing the request. Admins choose among OpenAI GPT models, Anthropic Claude Sonnet 4 and Opus 4.1, any model in the Azure AI Model Catalog, or a bring-your-own-model connection to an Azure AI Foundry deployment, but Sim's own model-fallback only retries the same model with hosted keys, not a cross-model fallback like this.",
       shortDescription:
-        'OpenAI, Anthropic Claude, Azure AI Model Catalog, or a bring-your-own model.',
+        'Falls back to the default OpenAI model automatically if a selected model is disabled.',
       source: {
         url: 'https://www.microsoft.com/en-us/microsoft-copilot/blog/copilot-studio/anthropic-joins-the-multi-model-lineup-in-microsoft-copilot-studio/',
         label: 'Anthropic joins the multi-model lineup in Microsoft Copilot Studio',
diff --git a/apps/sim/lib/compare/data/competitors/n8n.ts b/apps/sim/lib/compare/data/competitors/n8n.ts
index 75a56a31c48..cdfc3e3d410 100644
--- a/apps/sim/lib/compare/data/competitors/n8n.ts
+++ b/apps/sim/lib/compare/data/competitors/n8n.ts
@@ -104,11 +104,10 @@ export const n8nProfile: CompetitorProfile = {
       source: { url: 'https://trust.n8n.io/', label: 'n8n Trust Center', asOf: '2026-07-02' },
     },
     {
-      title:
-        'SSO/SAML/LDAP, audit logging, and dedicated SLA support gated to paid/Enterprise tiers',
+      title: 'No RBAC or workflow sharing at all on the free Community edition',
       description:
-        'SSO/SAML/LDAP, custom project roles, audit log export/SIEM streaming, and dedicated SLA-backed support are not available on the Community (free, self-hosted) edition; they require the Business or Enterprise plans.',
-      shortDescription: 'Governance features require Business or Enterprise, not the free tier.',
+        "n8n's free, self-hosted Community edition has zero role-based access control; project roles and any workflow sharing require a paid Business or Enterprise plan. Sim includes baseline workspace and organization roles (admin/write/read) on every plan, including the free tier, and reserves only SSO/SAML/OIDC, audit-log export, and SLA-backed support for Enterprise.",
+      shortDescription: 'Free Community edition has zero RBAC; Sim includes roles on every plan.',
       source: { url: 'https://n8n.io/pricing/', label: 'n8n Pricing', asOf: '2026-07-02' },
     },
     {
@@ -449,7 +448,7 @@ export const n8nProfile: CompetitorProfile = {
       evaluationGuardrails: {
         value: 'Yes: dedicated Evaluations feature (Light + Metric-based)',
         detail:
-          "n8n provides an Evaluation node/trigger and an Evaluations tab supporting 'Light evaluations' (manual test cases during development) and 'Metric-based evaluations' (scoring at scale for production), with built-in metrics (AI-judged Helpfulness, string similarity, categorization, tools-used) plus custom metrics. n8n's own 2026 AI Agent Development Tools report scores n8n 0 out of 2 on both \"JSON validity\" and \"Format check\" evaluation types, versus Sim's 2 out of 2 on both, via Sim's Guardrails block.",
+          "n8n provides an Evaluation node/trigger and an Evaluations tab supporting 'Light evaluations' (manual test cases during development) and 'Metric-based evaluations' (scoring at scale for production), with built-in metrics (AI-judged Helpfulness, string similarity, categorization, tools-used) plus custom metrics.",
         shortValue: 'Light and metric-based evaluation testing',
         confidence: 'verified',
         sources: [
@@ -463,12 +462,6 @@ export const n8nProfile: CompetitorProfile = {
             label: 'n8n Blog: Introducing Evaluations for AI workflows',
             asOf: '2026-07-02',
           },
-          {
-            url: 'https://n8n.io/reports/2026-ai-agent-development-tools/#vendors',
-            label:
-              'n8n: 2026 AI Agent Development Tools report (JSON validity, Format check scores)',
-            asOf: '2026-07-02',
-          },
         ],
       },
       humanInTheLoop: {
@@ -533,7 +526,7 @@ export const n8nProfile: CompetitorProfile = {
           'No: n8n has no first-class, named reusable prompt/knowledge-snippet object that agents reference. Reuse is achieved informally by exporting/importing workflow JSON or calling a sub-workflow (e.g. a "Tool (Workflow)" node) as a reusable scratchpad, not by a dedicated skills library.',
         shortValue: 'No dedicated reusable skill/snippet object',
         detail:
-          'System prompts are configured per AI Agent node; the closest analog is reusable sub-workflows or exported JSON, not a named, invokable skill library. n8n\'s own 2026 AI Agent Development Tools report independently scores n8n 0 out of 2 on "Agent skills directory," versus Sim\'s 2 out of 2.',
+          'System prompts are configured per AI Agent node; the closest analog is reusable sub-workflows or exported JSON, not a named, invokable skill library.',
         confidence: 'verified',
         sources: [
           {
@@ -546,11 +539,6 @@ export const n8nProfile: CompetitorProfile = {
             label: 'Reusable thinking tools workflow template',
             asOf: '2026-07-02',
           },
-          {
-            url: 'https://n8n.io/reports/2026-ai-agent-development-tools/#vendors',
-            label: 'n8n: 2026 AI Agent Development Tools report (Agent skills directory score)',
-            asOf: '2026-07-02',
-          },
         ],
       },
       nativeChatDeployment: {
@@ -688,7 +676,7 @@ export const n8nProfile: CompetitorProfile = {
       customCodeSteps: {
         value: 'Yes: JavaScript and Python via Code node, plus a Custom Code Tool for AI agents',
         detail:
-          "n8n's Code node supports both JavaScript and Python for custom logic inside a workflow. A separate Custom Code Tool node lets an AI Agent call arbitrary code as one of its tools. n8n's own 2026 AI Agent Development Tools report scores n8n 1 out of 2 on \"Sandboxing,\" versus Sim's 2 out of 2, backed by Sim's isolated-vm (V8 isolate) sandbox running in a separate child process.",
+          "n8n's Code node supports both JavaScript and Python for custom logic inside a workflow. A separate Custom Code Tool node lets an AI Agent call arbitrary code as one of its tools.",
         shortValue: 'JavaScript and Python via Code node',
         confidence: 'estimated',
         sources: [
@@ -697,11 +685,6 @@ export const n8nProfile: CompetitorProfile = {
             label: 'n8n Tools Agent docs (mentions Custom Code Tool)',
             asOf: '2026-07-02',
           },
-          {
-            url: 'https://n8n.io/reports/2026-ai-agent-development-tools/#vendors',
-            label: 'n8n: 2026 AI Agent Development Tools report (Sandboxing score)',
-            asOf: '2026-07-02',
-          },
         ],
       },
       apiPublishing: {
diff --git a/apps/sim/lib/compare/data/competitors/openai-agentkit.ts b/apps/sim/lib/compare/data/competitors/openai-agentkit.ts
index eaf14bef792..036fb22767f 100644
--- a/apps/sim/lib/compare/data/competitors/openai-agentkit.ts
+++ b/apps/sim/lib/compare/data/competitors/openai-agentkit.ts
@@ -33,9 +33,9 @@ export const openaiAgentkitProfile: CompetitorProfile = {
     "OpenAI AgentKit bundled a visual Agent Builder, ChatKit embeddable chat UI, Connector Registry, Guardrails, and Evals for building agentic workflows on OpenAI's models. But OpenAI is winding down Agent Builder and Evals, with full shutdown November 30, 2026, in favor of the code-first Agents SDK or ChatGPT Workspace Agents.",
   standoutFeatures: [
     {
-      title: 'Code-first, all-OpenAI stack with no visual builder going forward',
+      title: 'An official, open-source Agents SDK wired natively to its own models',
       description:
-        "With Agent Builder and Evals winding down (full shutdown November 30, 2026), OpenAI's path forward is the code-first Agents SDK, openai-agents-python, open source under the MIT license with over 27,500 GitHub stars, natively wired into OpenAI's own model lineup. A team fully committed to an all-OpenAI, code-first stack, with no visual builder layer, gets that directly.",
+        "The Agents SDK, openai-agents-python, is open source under the MIT license with over 27,500 GitHub stars and natively wired into OpenAI's own model lineup. It's the path OpenAI is steering AgentKit users toward as Agent Builder and Evals wind down (full shutdown November 30, 2026), and a team fully committed to an all-OpenAI, code-first stack gets that directly.",
       shortDescription: 'Open-source code-first framework, natively wired to OpenAI models.',
       source: {
         url: 'https://github.com/openai/openai-agents-python',
@@ -66,13 +66,13 @@ export const openaiAgentkitProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Guardrails open-source safety layer',
+      title: 'Guardrails ships as a standalone, installable open-source package',
       description:
-        'Agent Builder shipped an open-source, modular guardrails layer that can mask/flag PII, detect jailbreaks, and apply other safety checks around agent behavior.',
-      shortDescription: 'Open-source guardrails layer masks PII and detects jailbreaks.',
+        "Agent Builder's Guardrails node is backed by a separately installable, open-source package that can be used outside AgentKit entirely, adding jailbreak detection alongside PII masking and other safety checks. Sim's Guardrails block covers PII masking and hallucination/RAG scoring too, but it's a built-in workflow block, not a standalone library you can drop into an unrelated codebase.",
+      shortDescription: 'Standalone open-source package adds jailbreak detection to PII masking.',
       source: {
-        url: 'https://openai.com/index/introducing-agentkit/',
-        label: 'Introducing AgentKit (via search excerpt)',
+        url: 'https://guardrails.openai.com/',
+        label: 'OpenAI Guardrails',
         asOf: '2026-07-02',
       },
     },
@@ -105,7 +105,7 @@ export const openaiAgentkitProfile: CompetitorProfile = {
       title:
         'Usage-based, per-token/per-call pricing with no published flat plan for AgentKit itself',
       description:
-        'There is no dedicated AgentKit subscription tier. Costs are the sum of model tokens, Code Interpreter sessions ($0.03-$1.92 per session by memory tier), and File Search ($0.10/GB-day storage plus $2.50 per 1,000 tool calls), which makes cost forecasting harder than a flat, seat-based plan.',
+        "There is no dedicated AgentKit subscription tier. Costs are the sum of model tokens, Code Interpreter sessions ($0.03-$1.92 per session by memory tier), and File Search ($0.10/GB-day storage plus $2.50 per 1,000 tool calls), which makes cost forecasting harder than Sim's published pricing, where a Pro seat is a flat $25/user/month with a monthly credit allowance included, rather than open-ended per-token and per-call metering.",
       shortDescription: 'No flat plan; costs scale with token and tool usage.',
       source: {
         url: 'https://developers.openai.com/api/docs/pricing',
diff --git a/apps/sim/lib/compare/data/competitors/openclaw.ts b/apps/sim/lib/compare/data/competitors/openclaw.ts
index ec108c84c4f..e00e959c956 100644
--- a/apps/sim/lib/compare/data/competitors/openclaw.ts
+++ b/apps/sim/lib/compare/data/competitors/openclaw.ts
@@ -32,7 +32,7 @@ export const openClawProfile: CompetitorProfile = {
     {
       title: 'ClawHub Skills marketplace with multi-scanner security pipeline',
       description:
-        'Skills are Markdown-based instruction packages (SKILL.md files) installable from the public ClawHub registry, git repos, or local directories. Every published skill runs through a ClawScan pipeline combining static analysis, VirusTotal, and NVIDIA SkillSpector (added June 2026), and gets a Clean/Suspicious/Malicious verdict plus a Skill Card documenting provenance.',
+        "Skills are Markdown-based instruction packages (SKILL.md files) installable from the public ClawHub registry, git repos, or local directories. Every published skill runs through a ClawScan pipeline combining static analysis, VirusTotal, and NVIDIA SkillSpector (added June 2026), and gets a Clean/Suspicious/Malicious verdict plus a Skill Card documenting provenance. This scanning was introduced only after researchers found roughly 900 ClawHub skills with a documented credential-leak or malware finding (see limitations), and OpenClaw's own docs still tell users to treat third-party skills as untrusted code.",
       shortDescription:
         'Markdown skills from ClawHub, each scanned by static analysis, VirusTotal, and SkillSpector.',
       source: {
@@ -66,11 +66,11 @@ export const openClawProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Markdown-file memory instead of an opaque vector store',
+      title: 'Markdown-file memory with no ingestion pipeline',
       description:
-        "Long-term memory is stored as plain, human-readable Markdown files (daily notes plus a curated MEMORY.md), layered with semantic search (memorySearch), instead of hiding retrieved context inside a vector database the user can't inspect or edit directly.",
+        'Long-term memory is stored as plain, human-readable Markdown files (daily notes plus a curated MEMORY.md), layered with semantic search (memorySearch), instead of running an ingestion/chunking pipeline into a database. The files are natively git-trackable and editable in any text editor, with no separate KB module standing between the user and their own memory.',
       shortDescription:
-        'Long-term memory lives in editable Markdown files, not a hidden vector store.',
+        'Long-term memory lives in plain, git-trackable Markdown files, not a database-backed KB module.',
       source: {
         url: 'https://docs.openclaw.ai/concepts/memory',
         label: 'OpenClaw Docs: Memory overview',
@@ -128,11 +128,10 @@ export const openClawProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'No enterprise compliance certifications (SOC 2, ISO 27001, HIPAA)',
+      title: 'No SOC 2 report or other compliance attestation',
       description:
-        'OpenClaw is a self-hosted open-source project run by a non-profit foundation, not a vendor selling a hosted service, so it publishes no SOC 2, ISO 27001, HIPAA, or similar compliance attestation. Security for data-at-rest and processing falls entirely on the operator running their own instance.',
-      shortDescription:
-        'No SOC 2/ISO/HIPAA attestations; the self-hosting operator owns all compliance risk.',
+        'OpenClaw is a self-hosted open-source project run by a non-profit foundation, not a vendor selling a hosted service, so it publishes no SOC 2 report or other compliance attestation. Sim is SOC 2 compliant; like OpenClaw, Sim does not currently hold ISO 27001 or HIPAA certification. Security for data-at-rest and processing on OpenClaw falls entirely on the operator running their own instance.',
+      shortDescription: 'No SOC 2 report; the self-hosting operator owns all compliance risk.',
       source: {
         url: 'https://docs.openclaw.ai/gateway/security',
         label: 'OpenClaw Docs: Security',
diff --git a/apps/sim/lib/compare/data/competitors/pipedream.ts b/apps/sim/lib/compare/data/competitors/pipedream.ts
index fed13bb7424..ee506a3994a 100644
--- a/apps/sim/lib/compare/data/competitors/pipedream.ts
+++ b/apps/sim/lib/compare/data/competitors/pipedream.ts
@@ -38,17 +38,6 @@ export const pipedreamProfile: CompetitorProfile = {
       shortDescription: 'Every workflow step can drop into custom code in four languages.',
       source: { url: 'https://pipedream.com/docs', label: 'Pipedream Docs', asOf: '2026-07-02' },
     },
-    {
-      title: 'Edit workflows with natural language',
-      description:
-        "An 'Edit with AI' button in the workflow builder header or any code step lets users modify an existing workflow using natural-language instructions.",
-      shortDescription: "An 'Edit with AI' button lets users modify workflows by prompt.",
-      source: {
-        url: 'https://pipedream.com/blog/build-workflows-faster-with-ai/',
-        label: 'Pipedream Blog: Build workflows faster with AI',
-        asOf: '2026-07-04',
-      },
-    },
     {
       title: 'Source-available component registry on GitHub',
       description:
@@ -96,13 +85,13 @@ export const pipedreamProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Support response-time SLAs not publicly documented',
+      title: 'No native fail-and-continue branch on step errors',
       description:
-        'The Enterprise plan reportedly includes a dedicated Success Engineer and a platform uptime guarantee, but support response-time SLAs are not published on any plan page and appear to be negotiated directly with sales.',
-      shortDescription: 'No published response-time SLA on any public plan page.',
+        'By default an unhandled step error halts the entire workflow execution. Auto-retry (Advanced plan and above) and the global $error event stream let teams react to a failure after the fact, but neither lets the original run continue past the failing step in the same execution without hand-coded try/catch or conditional logic.',
+      shortDescription: 'An unhandled step error halts the whole run by default.',
       source: {
-        url: 'https://pipedream.com/pricing',
-        label: 'Pipedream Pricing page',
+        url: 'https://pipedream.com/docs/workflows/building-workflows/errors',
+        label: 'Pipedream Docs: Handling Errors',
         asOf: '2026-07-02',
       },
     },
@@ -1099,21 +1088,21 @@ export const pipedreamProfile: CompetitorProfile = {
       },
       companyMaturity: {
         value:
-          'Founded 2019, San Francisco; ~11-50 employees pre-acquisition; raised ~$22M across 2 rounds; acquired by Workday (deal announced Nov 19, 2025)',
+          'Founded 2019, San Francisco; ~11-50 employees pre-acquisition; raised ~$22M across 2 rounds; agreed to be acquired by Workday (signed Nov 19, 2025), with no public confirmation the deal has closed as of mid-2026',
         detail:
-          "Per Crunchbase, Pipedream was founded in 2019 and headquartered in San Francisco, CA, with founders including Tod Sacerdoti (CEO), Dylan Sather, TJ Koblentz, and Pravin Savkar; it raised a total of ~$22M across 2 funding rounds (investors include Felicis and CRV) and had a headcount signal of 11-50 employees. Workday announced a definitive agreement to acquire Pipedream on November 19, 2025, with the transaction expected to close in Workday's Q4 FY2026 (by end of January 2026); Pipedream is now being positioned as Workday's integration layer for AI agent workflows across Workday and third-party apps.",
-        shortValue: 'Founded 2019; acquired by Workday, Nov 2025',
-        confidence: 'verified',
+          "Per Crunchbase, Pipedream was founded in 2019 and headquartered in San Francisco, CA, with founders including Tod Sacerdoti (CEO), Dylan Sather, TJ Koblentz, and Pravin Savkar; it raised a total of ~$22M across 2 funding rounds (investors include Felicis and CRV) and had a headcount signal of 11-50 employees. Workday signed a definitive agreement to acquire Pipedream on November 19, 2025, with the transaction originally expected to close in Workday's Q4 FY2026 (by end of January 2026), subject to closing conditions. That expected close date has since passed, and as of this writing neither Workday nor Pipedream has publicly announced that the deal has closed.",
+        shortValue: 'Founded 2019; agreed to be acquired by Workday, close unconfirmed',
+        confidence: 'estimated',
         sources: [
           {
             url: 'https://newsroom.workday.com/2025-11-19-Workday-Signs-Definitive-Agreement-to-Acquire-Pipedream',
             label: 'Workday Newsroom – Workday Signs Definitive Agreement to Acquire Pipedream',
-            asOf: '2026-07-02',
+            asOf: '2026-07-06',
           },
           {
             url: 'https://pipedream.com/blog/pipedream-to-be-acquired-by-workday/',
             label: 'Pipedream Blog – Pipedream to be acquired by Workday',
-            asOf: '2026-07-02',
+            asOf: '2026-07-06',
           },
           {
             url: 'https://www.crunchbase.com/organization/pipedream',
diff --git a/apps/sim/lib/compare/data/competitors/power-automate.ts b/apps/sim/lib/compare/data/competitors/power-automate.ts
index 52c6ecc1654..84016f60493 100644
--- a/apps/sim/lib/compare/data/competitors/power-automate.ts
+++ b/apps/sim/lib/compare/data/competitors/power-automate.ts
@@ -29,7 +29,7 @@ export const powerAutomateProfile: CompetitorProfile = {
     {
       title: 'Solutions-based ALM with managed/unmanaged packages and pipelines',
       description:
-        'Flows package into Dataverse Solutions (managed vs. unmanaged) and promote from dev to test to production via Power Platform Pipelines, with environment variables swapping per-environment references, a full multi-environment promotion model, not just single-workflow versioning.',
+        'Flows package into Dataverse Solutions (managed vs. unmanaged) and promote from dev to test to production via Power Platform Pipelines, with environment variables swapping per-environment references. The promotion is tied to Dataverse as a shared enterprise data platform, with centralized admin governance over which environments a solution can move into.',
       shortDescription:
         'Flows promote dev to test to production via Dataverse Solutions and Pipelines.',
       source: {
@@ -50,11 +50,11 @@ export const powerAutomateProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Multi-model choice including Anthropic Claude in Copilot Studio',
+      title: 'Tenant-wide admin gating with automatic model fallback in Copilot Studio',
       description:
-        'Copilot Studio added Claude Sonnet 4 and Claude Opus 4.1 as selectable models alongside OpenAI GPT models, with admin controls to enable/restrict and automatic fallback to the default OpenAI model if disabled.',
+        'Copilot Studio lets admins enable or restrict which models (including Anthropic Claude Sonnet 4 and Opus 4.1 alongside OpenAI GPT models) are available tenant-wide from the Microsoft 365 Admin Center, and agents automatically fall back to the default OpenAI GPT-4o model if their selected model is disabled, with no additional configuration required.',
       shortDescription:
-        'Copilot Studio supports Claude Sonnet 4 and Opus 4.1 alongside OpenAI models.',
+        'Admins gate model access tenant-wide, with automatic fallback to the default model.',
       source: {
         url: 'https://www.microsoft.com/en-us/microsoft-copilot/blog/copilot-studio/anthropic-joins-the-multi-model-lineup-in-microsoft-copilot-studio/',
         label: 'Anthropic joins the multi-model lineup in Microsoft Copilot Studio',
@@ -76,13 +76,13 @@ export const powerAutomateProfile: CompetitorProfile = {
   ],
   limitations: [
     {
-      title: 'No native side-by-side version diff/compare view',
+      title: 'No live, concurrent multi-user editing in the flow designer',
       description:
-        'Power Automate has version history with restore, but no built-in visual diff between two flow versions. Teams export both definitions and run a manual text diff in an external tool like VS Code to see what changed.',
-      shortDescription: 'No built-in visual diff between two flow versions.',
+        "Power Automate's cloud flow designer supports sharing a flow with co-owners, but not live, concurrent multi-user editing with visible cursors and synced changes on the same flow. Microsoft's live coauthoring feature exists for Power Apps Studio canvas apps, a separate product, not the Power Automate flow designer.",
+      shortDescription: 'No true live co-editing in the flow designer, only owner-level sharing.',
       source: {
-        url: 'https://sharepains.com/2024/04/26/version-history-in-power-automate-flows/',
-        label: 'Version history in Power Automate flows',
+        url: 'https://learn.microsoft.com/en-us/power-automate/guide-to-cloud-flow-sharing-permissions',
+        label: 'Guide to cloud flow sharing and permissions | Microsoft Learn',
         asOf: '2026-07-02',
       },
     },
diff --git a/apps/sim/lib/compare/data/competitors/retool.ts b/apps/sim/lib/compare/data/competitors/retool.ts
index 8133fd51c86..e9f6b208d36 100644
--- a/apps/sim/lib/compare/data/competitors/retool.ts
+++ b/apps/sim/lib/compare/data/competitors/retool.ts
@@ -39,7 +39,7 @@ export const retoolProfile: CompetitorProfile = {
     {
       title: 'Retool Vectors (managed vector store)',
       description:
-        "A Retool-managed vector database that automatically indexes uploaded text, PDFs, or web pages, so AI apps and agents can look up relevant content with one click instead of building a custom search pipeline. The lookups always run through OpenAI's embedding API, even when the chat model is a different provider.",
+        "A Retool-managed vector database that automatically indexes uploaded text, PDFs, or web pages, so AI apps and agents can look up relevant content with one click. The lookups always run through OpenAI's embedding API, even when the chat model is a different provider, so there is no way to tune or swap the embedding step independently of the chat model.",
       shortDescription:
         'Managed vector database with automatic indexing for one-click content lookup.',
       source: {
@@ -49,11 +49,11 @@ export const retoolProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Bidirectional MCP support',
+      title: 'Workspace-level MCP server for platform management',
       description:
-        'Retool Agents can connect outbound to external MCP servers (a standard for plugging AI agents into outside tools) to pull in tools like GitHub or Jira. Retool also exposes its own workspace as an MCP server (public beta), so build and management actions, such as creating apps, running queries, and managing users, can be performed directly from Claude, Cursor, Codex, or Kiro. This does not let you publish an individual deployed app or workflow as its own standalone MCP tool for outside consumption.',
+        'Retool Agents can connect outbound to external MCP servers (a standard for plugging AI agents into outside tools) to pull in tools like GitHub or Jira. Retool also exposes its own workspace as an MCP server (public beta), so platform-administration actions, such as creating apps, running queries, and managing users, can be performed directly from Claude, Cursor, Codex, or Kiro. This does not let you publish an individual deployed app or workflow as its own standalone MCP tool for outside consumption.',
       shortDescription:
-        'Connects to external MCP servers and exposes workspace management actions as one.',
+        'Connects to external MCP servers and exposes workspace administration actions as one.',
       source: {
         url: 'https://retool.com/blog/how-to-use-mcp-in-retool',
         label: 'How to use MCP in Retool',
@@ -63,8 +63,8 @@ export const retoolProfile: CompetitorProfile = {
     {
       title: 'Retool Agents (deterministic + non-deterministic decisioning)',
       description:
-        'A dedicated agent-building surface that combines code-based deterministic logic, LLM-based non-deterministic decisions, and human-in-the-loop steps within one automation, distinct from the classic Workflows product.',
-      shortDescription: 'Combines deterministic logic, LLM decisions, and human-in-the-loop steps.',
+        'A dedicated agent-building surface, separate from the classic Workflows product, that combines code-based deterministic logic, LLM-based non-deterministic decisions, and human-in-the-loop steps within one automation. Sim covers the same ground with agent, function, and human-in-the-loop blocks combined directly in its single workflow builder, without needing a second product.',
+      shortDescription: 'A separate product from classic Workflows for combining these primitives.',
       source: {
         url: 'https://docs.retool.com/agents',
         label: 'Retool Agents docs',
diff --git a/apps/sim/lib/compare/data/competitors/stackai.ts b/apps/sim/lib/compare/data/competitors/stackai.ts
index b3634226c5f..7a4bf12253b 100644
--- a/apps/sim/lib/compare/data/competitors/stackai.ts
+++ b/apps/sim/lib/compare/data/competitors/stackai.ts
@@ -17,10 +17,10 @@ export const stackaiProfile: CompetitorProfile = {
     'StackAI is a proprietary, enterprise-focused visual platform for building, deploying, and governing AI agents, connecting LLMs and business systems through a drag-and-drop, low-code node builder.',
   standoutFeatures: [
     {
-      title: 'SOC 2 Type II and ISO 27001 certified, with a public Trust Center',
+      title: 'ISO 27001 certified, with a public Trust Center detailing pen tests and DPAs',
       description:
-        'StackAI publishes a Trust Center (trust.stackai.com) documenting SOC 2 Type II and ISO 27001 certification, third-party penetration test results, and DPAs with OpenAI and Anthropic.',
-      shortDescription: 'Public Trust Center with SOC 2, ISO 27001, and pen test results.',
+        'StackAI publishes a Trust Center (trust.stackai.com) documenting ISO 27001 certification, third-party penetration test results, and DPAs with OpenAI and Anthropic. StackAI also holds a SOC 2 Type II audit, but so does Sim, so ISO 27001 is the actual point of difference here.',
+      shortDescription: 'Public Trust Center with ISO 27001, pen test results, and DPAs.',
       source: {
         url: 'https://trust.stackai.com/',
         label: 'StackAI Trust Center',
@@ -28,9 +28,9 @@ export const stackaiProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Agentic Development Life Cycle (dev/staging/production promotion)',
+      title: 'PR-gated approval workflow for promoting agents between environments',
       description:
-        'StackAI provides three default, isolated environments (development, staging, production), plus custom environments. Promotion between them is gated by pull requests, each environment can connect to its own data sources, and an admin approval queue sits before production deploys.',
+        "StackAI provides three default, isolated environments (development, staging, production), plus custom environments. Promotion between them requires a pull request that must be reviewed and approved, with an admin approval queue sitting before production deploys. Sim also supports forking a workspace into dev/qa/prod-style environments with diff and promote/rollback, but without a mandatory PR-review or approval gate, and that capability is itself gated to Sim's Enterprise plan on hosted Sim (or a feature flag on self-hosted deployments).",
       shortDescription: 'PR-gated dev/staging/production promotion with admin approval queues.',
       source: {
         url: 'https://www.stackai.com/blog/the-agentic-development-life-cycle-how-to-manage-ai-agents-at-scale',
@@ -39,9 +39,9 @@ export const stackaiProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Version history with diff/compare and rollback',
+      title: 'Full version diff/compare on every manual edit, not just AI-generated changes',
       description:
-        'Every save creates a full version snapshot of an agent. A comparison tool shows added or removed nodes, prompt and LLM config changes, and connection changes. Any version can be reverted, and reverting creates a new version rather than erasing history.',
+        'Every save creates a full version snapshot of an agent, regardless of whether the change was made manually or by an AI assistant. A comparison tool shows added or removed nodes, prompt and LLM config changes, and connection changes. Any version can be reverted, and reverting creates a new version rather than erasing history. Sim diffs and reverts Copilot-generated edits, but manual edits only get local undo/redo, not a versioned diff.',
       shortDescription: 'Full version snapshots with diff/compare and one-click rollback.',
       source: {
         url: 'https://www.stackai.com/blog/the-agentic-development-life-cycle-how-to-manage-ai-agents-at-scale',
@@ -49,17 +49,6 @@ export const stackaiProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
-    {
-      title: 'Human-in-the-loop approval gating before side effects',
-      description:
-        'A workflow can pause at a decision point and send an approval request, for example via Slack, Teams, or email, before a risky action like sending an email, writing to a database, or provisioning access. The run resumes once a human approves, rejects, or gives feedback.',
-      shortDescription: 'Pauses workflows for human approval before risky tool calls execute.',
-      source: {
-        url: 'https://www.stackai.com/blog/introducing-stackai-human-in-the-loop-agentic-workflows-you-can-trust',
-        label: 'Introducing StackAI Human-in-the-Loop - StackAI blog',
-        asOf: '2026-07-02',
-      },
-    },
     {
       title: 'On-prem / VPC self-hosted deployment for enterprise',
       description:
diff --git a/apps/sim/lib/compare/data/competitors/tines.ts b/apps/sim/lib/compare/data/competitors/tines.ts
index eab44d28cbf..00437900da7 100644
--- a/apps/sim/lib/compare/data/competitors/tines.ts
+++ b/apps/sim/lib/compare/data/competitors/tines.ts
@@ -16,18 +16,6 @@ export const tinesProfile: CompetitorProfile = {
   oneLiner:
     'Tines is a proprietary workflow automation platform, available cloud-hosted or self-hosted, originally built for security operations. Teams build event-driven "Stories" via a visual no/low-code canvas, natural language, or the API. It recently added native AI agent, MCP, and copilot capabilities.',
   standoutFeatures: [
-    {
-      title: 'API-centric integration model',
-      description:
-        'Instead of a fixed library of app connectors, Tines is built around a generic HTTP Request action that calls any API directly, trading pre-built connectors for broader reach and more manual setup.',
-      shortDescription:
-        'A generic HTTP Request action reaches any API instead of fixed connectors.',
-      source: {
-        url: 'https://www.tines.com/blog/solving-the-integrations-problem/',
-        label: 'Solving the integrations problem',
-        asOf: '2026-07-02',
-      },
-    },
     {
       title: 'ISO 42001 AI-governance certification',
       description:
@@ -40,24 +28,25 @@ export const tinesProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Workbench natural-language builder',
+      title: 'API-centric integration model',
       description:
-        'Users describe an automation in plain language and Workbench, the platform-wide AI assistant, generates the Story automatically. Workbench absorbed the former "Story Copilot," which was renamed "Workbench for Storyboard" on June 2, 2026.',
-      shortDescription: 'Workbench turns plain-language descriptions into working automations.',
+        'Instead of a fixed library of app connectors, Tines is built around a generic HTTP Request action that calls any API directly, trading pre-built connectors for broader reach and more manual setup.',
+      shortDescription:
+        'A generic HTTP Request action reaches any API instead of fixed connectors.',
       source: {
-        url: 'https://www.tines.com/whats-new/workbench-for-stories/',
-        label: "Tines What's New",
-        asOf: '2026-07-04',
+        url: 'https://www.tines.com/blog/solving-the-integrations-problem/',
+        label: 'Solving the integrations problem',
+        asOf: '2026-07-02',
       },
     },
     {
-      title: 'Native MCP server and client support',
+      title: 'Tines University certification ladder',
       description:
-        'Tines can act as both an MCP server (exposing its actions to AI clients) and an MCP client (consuming external MCP tools), framed around governed, policy-aligned AI access rather than unrestricted tool use.',
-      shortDescription: 'Acts as both an MCP server and client for governed AI tool access.',
+        'Tines University pairs free foundational courses with instructor-led and self-paced Bootcamps and two certification tiers, Core and Advanced, that builders can share on LinkedIn. Sim Academy is a structured docs section without a formal certification path.',
+      shortDescription: 'Core and Advanced certifications builders can share on LinkedIn.',
       source: {
-        url: 'https://www.tines.com/platform/ai/',
-        label: 'AI Agents, Copilots & MCP | Tines',
+        url: 'https://www.tines.com/get-certified/',
+        label: 'Get certified | Tines',
         asOf: '2026-07-02',
       },
     },
@@ -1131,6 +1120,8 @@ export const tinesProfile: CompetitorProfile = {
       companyMaturity: {
         value:
           'Founded 2018 (Dublin/Boston) by Eoin Hinchy and Thomas Kinsella; raised ~$272M total across 6 rounds, most recently a $125M Series C (Feb 2025) led by Goldman Sachs at unicorn valuation (~$1.125B); reported headcount roughly 500-550 as of early-to-mid 2026',
+        detail:
+          "No funding round beyond the Feb 2025 Series C is publicly confirmed as of this profile's research date; headcount reflects the most recently reported figures (around 548 employees as of March 2026), not necessarily the current count.",
         shortValue: 'Founded 2018, ~$272M raised, ~500-550 employees',
         confidence: 'estimated',
         sources: [
diff --git a/apps/sim/lib/compare/data/competitors/vellum.ts b/apps/sim/lib/compare/data/competitors/vellum.ts
index 5a68bdbb8f4..f09de0f34f4 100644
--- a/apps/sim/lib/compare/data/competitors/vellum.ts
+++ b/apps/sim/lib/compare/data/competitors/vellum.ts
@@ -18,10 +18,10 @@ export const vellumProfile: CompetitorProfile = {
     'Vellum is an enterprise AI development platform for building, evaluating, and deploying LLM prompts, workflows, and agents.',
   standoutFeatures: [
     {
-      title: 'SOC 2 Type 2 and HIPAA compliance with BAA',
+      title: 'HIPAA compliance with a signable BAA',
       description:
-        'Vellum has SOC 2 Type 2 attestation and HIPAA compliance. Enterprise customers can sign a Business Associate Agreement for handling protected health information, corroborated by a third-party Drata case study.',
-      shortDescription: 'SOC 2 Type 2 and HIPAA compliance with a signable BAA.',
+        'Vellum is HIPAA compliant and enterprise customers can sign a Business Associate Agreement for handling protected health information, corroborated by a third-party Drata case study. Sim does not currently offer HIPAA compliance (both platforms hold SOC 2 Type 2).',
+      shortDescription: 'HIPAA compliant with a signable BAA; Sim does not offer HIPAA.',
       source: {
         url: 'https://drata.com/customers/vellum',
         label: 'Vellum Case Study: Drata',
@@ -29,38 +29,16 @@ export const vellumProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Self-hosted / VPC enterprise deployment',
+      title: 'Vendor-managed dedicated VPC deployment',
       description:
-        "Enterprise customers can run the platform inside their own AWS, Azure, or GCP VPC (or on-prem) via a Replicated-based install, keeping prompts and documents inside the customer's network perimeter.",
-      shortDescription: 'Runs inside your own AWS/Azure/GCP VPC or on-prem.',
+        'Enterprise customers can get a Vellum-managed dedicated deployment inside a single-tenant AWS, Azure, or GCP VPC via a Replicated-based install. Sim offers self-hosting (Docker Compose or Kubernetes/Helm) but has no documented managed single-tenant/VPC hosting tier.',
+      shortDescription: 'Vendor-managed single-tenant VPC tier that Sim does not offer.',
       source: {
         url: 'https://docs.vellum.ai/self-hosting/getting-started/introduction',
         label: 'Self-Hosted Vellum: Vellum Docs',
         asOf: '2026-07-02',
       },
     },
-    {
-      title: "Natural-language agent building ('Vellum for Agents')",
-      description:
-        'Non-technical users can describe a goal in plain language and have Vellum generate a working agent, automatically handling model selection, prompting, and integration wiring.',
-      shortDescription: 'Describe a goal in plain language and Vellum builds the agent.',
-      source: {
-        url: 'https://www.vellum.ai/blog/introducing-vellum-for-agents',
-        label: 'Introducing Vellum for Agents',
-        asOf: '2026-07-02',
-      },
-    },
-    {
-      title: '$20M Series A, followed by a consumer pivot',
-      description:
-        "Vellum raised a $20M Series A in July 2025 (on top of a 2023 YC seed) to grow its enterprise platform. Since then, the company has separately launched a rebranded 'Personal Intelligence' consumer assistant product, open-sourced under MIT license on GitHub.",
-      shortDescription: '$20M Series A, then a separate consumer product launch.',
-      source: {
-        url: 'https://www.vellum.ai/blog/announcing-our-20m-series-a',
-        label: 'Announcing our $20m Series A: Vellum',
-        asOf: '2026-07-02',
-      },
-    },
   ],
   limitations: [
     {
@@ -94,14 +72,14 @@ export const vellumProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Deprecated legacy workflow nodes',
+      title: 'No native spreadsheet-style data tables',
       description:
-        'As of the January 2026 changelog, Vellum deprecated Merge, Conditional, and Output Nodes from the workflow builder UI (replaced by Merge Strategy, Ports, and Workflow Outputs respectively). Existing workflows using them continue to run but new instances can no longer be created.',
-      shortDescription:
-        'Legacy Merge/Conditional/Output nodes retired in favor of newer equivalents.',
+        'No Vellum documentation describes a native, spreadsheet-like data-table object with persistent rows/columns, keyboard navigation, or per-row writes from workflows; the closest documented tabular handling is uploading CSV/XLS files and extracting or generating structured output within workflow nodes, unlike Sim, which has a built-in Tables feature.',
+      shortDescription: 'No documented table/spreadsheet feature; CSV upload and extraction only.',
       source: {
-        url: 'https://docs.vellum.ai/changelog/2026/2026-01',
-        label: 'Vellum Changelog: January 2026',
+        url: 'https://docs.vellum.ai/developers/workflows-sdk/tutorials/document-data-extraction',
+        label:
+          'Vellum Docs: Document Data Extraction tutorial (the closest documented tabular-data handling)',
         asOf: '2026-07-02',
       },
     },
diff --git a/apps/sim/lib/compare/data/competitors/workato.ts b/apps/sim/lib/compare/data/competitors/workato.ts
index 247a5b55316..8e9fd6a00a4 100644
--- a/apps/sim/lib/compare/data/competitors/workato.ts
+++ b/apps/sim/lib/compare/data/competitors/workato.ts
@@ -44,13 +44,13 @@ export const workatoProfile: CompetitorProfile = {
       },
     },
     {
-      title: 'Enterprise MCP server hosting',
+      title: "Pre-built departmental agents ('Genies')",
       description:
-        'Workato exposes existing recipes and workflows as MCP tools through pre-built and remote/cloud-hosted MCP servers, letting any MCP-compatible client (Claude, ChatGPT, Agent Studio) dynamically discover and call enterprise workflows as agent tools without custom integration code.',
-      shortDescription: 'Recipes and workflows exposed as MCP tools for any compatible client.',
+        'Agent Studio ships ready-made departmental agents (Genies) for IT, Sales, HR, Support, CX, and Marketing that customers can deploy and customize directly, alongside the option to build a custom agent from scratch.',
+      shortDescription: 'Ready-made departmental agents (Genies) for IT, Sales, HR, and more.',
       source: {
-        url: 'https://docs.workato.com/mcp.html',
-        label: 'MCP | Workato docs',
+        url: 'https://docs.workato.com/agentic/agentic.html',
+        label: 'Agentic | Workato docs',
         asOf: '2026-07-02',
       },
     },
@@ -69,8 +69,9 @@ export const workatoProfile: CompetitorProfile = {
     {
       title: 'Bring-Your-Own-LLM for Agent Studio',
       description:
-        "Customers can power Genies with their own OpenAI or Anthropic API credentials instead of Workato's managed model contracts, giving direct control over LLM cost and vendor choice for agent workloads.",
-      shortDescription: 'Genies can run on customer-supplied OpenAI or Anthropic API keys.',
+        "Customers can power Genies with their own OpenAI or Anthropic API credentials instead of Workato's managed model contracts, giving direct control over LLM cost and vendor choice for agent workloads. This BYOLLM option is scoped to two providers and to Agent Studio specifically; Sim's own BYOK support spans any provider across every block, exempts usage from metered credit caps, and automatically round-robins multiple stored keys for the same provider.",
+      shortDescription:
+        'Genies can run on customer-supplied OpenAI or Anthropic API keys, narrower in scope than Sim.',
       source: {
         url: 'https://www.workato.com/product-hub/changelog/bring-your-own-llm-byollm-support-for-agent-studio/',
         label: 'Bring Your Own LLM (BYOLLM) Support for Agent Studio | Workato Product Hub',
@@ -123,6 +124,17 @@ export const workatoProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
+    {
+      title: 'SSO limited to SAML, no confirmed OIDC',
+      description:
+        "Workato's SSO documentation covers SAML with just-in-time provisioning and SAML role sync; it does not mention OIDC anywhere, so there is no public confirmation Workato supports it. Sim supports both SAML 2.0 and OIDC single sign-on.",
+      shortDescription: 'Documented SSO is SAML-only, with no confirmed OIDC support.',
+      source: {
+        url: 'https://docs.workato.com/user-accounts-and-teams/single-sign-on.html',
+        label: 'Workato Docs: Enable Single Sign-On for a Workato Workspace',
+        asOf: '2026-07-02',
+      },
+    },
   ],
   facts: {
     platform: {
diff --git a/apps/sim/lib/compare/data/competitors/zapier.ts b/apps/sim/lib/compare/data/competitors/zapier.ts
index 0e5cb9f56b8..34fe0616d6f 100644
--- a/apps/sim/lib/compare/data/competitors/zapier.ts
+++ b/apps/sim/lib/compare/data/competitors/zapier.ts
@@ -50,7 +50,7 @@ export const zapierProfile: CompetitorProfile = {
     {
       title: 'Zapier Copilot (natural-language build assistant)',
       description:
-        'Copilot lets users describe an automation or agent in plain language and generates a draft Zap or agent, including custom code steps to fill integration gaps. In open beta.',
+        'Copilot (open beta) lets users describe an automation or agent in plain language and generates a draft Zap or agent, including custom code steps to fill integration gaps. Sim offers the same natural-language building capability via Chat and in-editor Copilot.',
       shortDescription: 'Describe an automation and Copilot builds the Zap or agent for you.',
       source: {
         url: 'https://zapier.com/blog/zapier-copilot-guide/',
@@ -58,17 +58,6 @@ export const zapierProfile: CompetitorProfile = {
         asOf: '2026-07-02',
       },
     },
-    {
-      title: 'Multi-provider LLM support across Zaps/Agents/Chatbots',
-      description:
-        'Zapier lets users choose among OpenAI (GPT), Anthropic (Claude), and Google (Gemini) models inside AI-powered steps and chatbots, with BYOK for OpenAI, Anthropic, Gemini, and Azure OpenAI keys.',
-      shortDescription: 'Choose OpenAI, Anthropic, or Google models, with BYOK support.',
-      source: {
-        url: 'https://zapier.com/blog/ai-models-on-zapier/',
-        label: 'Which AI models can you automate on Zapier?',
-        asOf: '2026-07-02',
-      },
-    },
   ],
   limitations: [
     {
@@ -85,8 +74,9 @@ export const zapierProfile: CompetitorProfile = {
     {
       title: 'Task-based pricing scales quickly with usage',
       description:
-        'Every Zap action step, and every MCP tool call at 2 tasks each, consumes a metered task, so pricing and plan tier are driven by execution volume rather than a flat seat or workflow count. Costs rise fast as usage grows.',
-      shortDescription: 'Costs scale with execution volume, not a flat seat count.',
+        "Every Zap action step, and every MCP tool call at 2 tasks each, consumes a metered task, so pricing and plan tier are driven by execution volume rather than a flat seat or workflow count. There's no way to fully exempt usage from the task count: even bringing your own AI provider key only covers AI-powered steps and chatbots, not the underlying app-action tasks. Costs rise fast as usage grows with no BYOK path around the per-task metering.",
+      shortDescription:
+        'Costs scale with task volume; BYOK only covers AI steps, not app-action tasks.',
       source: {
         url: 'https://www.activepieces.com/blog/zapier-pricing',
         label: 'Zapier Pricing Breakdown (third-party analysis)',
@@ -108,7 +98,7 @@ export const zapierProfile: CompetitorProfile = {
       title: 'No documented data-residency choice',
       description:
         "Zapier's infrastructure runs on AWS in the United States, with no selectable regional data residency or EU-only hosting option for standard customers.",
-      shortDescription: 'Runs on AWS in the US only, no regional hosting choice.',
+      shortDescription: 'No selectable region; standard customers are US-only (AWS).',
       source: {
         url: 'https://zapier.com/security-compliance',
         label: 'Zapier Security & Compliance (via search cache)',