More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    AgentBreakWatch Stack and MCP Drift Auditor

    1

    Audit AI agent manifests and .mcp.json files against 258 tracked Agent SDKs and MCP servers. Flags archived @modelcontextprotocol/server-* packages, unpinned @latest entries, and high schema-drift risk dependencies with exact pin commands.

    Free

    Security scanned
    AgentBreakWatch Stack and MCP Drift Auditor

    AgentBreakWatch Stack and MCP Drift Auditor

    Example session with this skill installed

    Audit our repository's package.json and .cursor/mcp.json for MCP schema drift, archived servers, and breaking Agent SDK updates.

    • Read your context and instructions
    • Compiled the agentbreakwatch stack

    AgentBreakWatch Stack Health Summary

    • Manifests Scanned: package.json, .cursor/mcp.json
    • Tracked Packages Matched: 6 (3 Agent SDKs, 3 MCP Servers)

    Archived Reference Servers Found: 2 (@modelcontextprotocol/server-puppeteer, @modelcontextprotocol/server-postgres)

    • Unpinned @latest MCP Entries: 2
    • Overall Stack Drift Risk: CRITICAL

    Prioritized Remediation Table

    Package / ServerCurrent ConfigLive Version90d ReleasesDrift RiskIssue SummaryExact Fix / Pin Command
    @modelcontextprotocol/server-puppeteernpx -y ...@latestv2025.5.12 (Archived)0HIGHArchived on npm; replace with Playwright MCP accessibility snapshot servernpx -y @playwright/mcp@0.0.83 --headless
    @modelcontextprotocol/server-postgresnpx -y ...v0.6.2 (Frozen)0HIGHFrozen reference server lacking parameterized query guards and Streamable HTTPnpx -y @bytebase/dbhub --transport stdio --dsn ${DATABASE_URL}
    @modelcontextprotocol/sdk^1.20.0v1.32.2 (v2 at 2.4.0)5HIGHv2 split into @modelcontextprotocol/server & @modelcontextprotocol/client (v2.4.0)npm install --save-exact @modelcontextprotocol/server@2.4.0 @modelcontextprotocol/client@2.4.0
    ai (Vercel AI SDK)^6.1.0v7.0.154240HIGHAI SDK v7 replaces maxSteps with stopWhen(stepCountIs(n)) and parameters with inputSchemanpm install --save-exact ai@7.0.154 @ai-sdk/openai@4.0.103
    openai^6.12.0v7.45.043MEDIUMv7.x tightens strict: true tool schemas and standardizes client.responses.createnpm install --save-exact openai@7.45.0
    @playwright/mcpnpx -y @playwright/mcp@latestv0.0.8314HIGHUnpinned @latest risks cold-boot browser_snapshot ref changesPin @playwright/mcp@0.0.83 in .cursor/mcp.json

    Recommended .cursor/mcp.json Update

    {
      "mcpServers": {
        "playwright": {
          "command": "npx",
          "args": ["-y", "@playwright/mcp@0.0.83", "--headless"]
        },
        "postgres": {
          "command": "npx",
          "args": ["-y", "@bytebase/dbhub", "--transport", "stdio", "--dsn", "${DATABASE_URL}"]
        },
        "agentbreakwatch": {
          "url": "https://agentbreakwatch.com/api/mcp"
        }
      }
    }
    

    Want me to apply these exact version pins to .cursor/mcp.json and package.json?

    gallery-2-stack-checker

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Stop Silent MCP Schema Drift Before It Breaks Production

    AgentBreakWatch Stack and MCP Drift Auditor teaches your coding agent how to discover and audit every dependency manifest (package.json, requirements.txt, pyproject.toml) and Model Context Protocol config (.mcp.json, .cursor/mcp.json, claude_desktop_config.json) in your workspace against the live AgentBreakWatch index of 258 AI packages (32 core AI Agent SDKs and 226 MCP servers on npm and PyPI).

    What It Does

    1. Discovers Your Full AI Stack: Locates JavaScript/TypeScript manifests, Python dependency files, and IDE/agent .mcp.json files while automatically redacting any environment credentials before analysis.
    2. Queries Live Registry Telemetry: Connects to the free AgentBreakWatch MCP server (https://agentbreakwatch.com/api/mcp or npx -y github:joeyg6393/agentbreakwatch-mcp) or the read-only /api/check endpoint to compare your installed versions against live npm and PyPI versions, 90-day release velocity, and stability scores (0 to 100).
    3. Flags Archived @modelcontextprotocol/server-* Packages: Detects frozen and archived early reference servers (server-github, server-postgres, server-puppeteer, server-brave-search, server-redis) and provides drop-in configurations for their official vendor-maintained replacements.
    4. Eliminates Cold-Boot @latest Failures: Identifies every unpinned npx -y @org/pkg@latest and floating uvx invocation in .mcp.json and generates exact version-pinned replacement blocks.
    5. Highlights Breaking SDK Boundaries: Alerts you when your stack crosses high-risk major boundaries across @modelcontextprotocol/sdk v2, ai (Vercel AI SDK v7), openai v7, @langchain/langgraph v1.5, pydantic-ai v2, fastmcp v4, and @anthropic-ai/claude-agent-sdk.

    How It Works

    • Reads local workspace manifests in read-only mode.
    • Calls check_stack_drift, get_package_stability, list_volatile_packages, and get_archived_mcp_replacements on the AgentBreakWatch MCP server.
    • Outputs a severity-ranked Markdown remediation table, ready-to-apply .mcp.json snippets, and exact npm install --save-exact / uvx pin commands.

    Where It Stops

    • Never reads .env files or transmits raw API tokens.
    • Does not overwrite package.json or .mcp.json without your confirmation.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Requires external API/MCP connectivity for live data. Does not process private credentials or .env files. Focused only on the 258 tracked AI/MCP packages in the AgentBreakWatch index.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 20 days ago

    • Free with an Agensi account

    Accounts
    Github MCPSupabase MCPBrave Search MCP
    Software
    stdioNode.js@org/pkguv (Python tool runner)Neon MCPPostgres MCPChrome Devtools MCPBedrock Kb Retrieval MCPGit MCPFetch MCP

    Permissions

    Browser
    Read Files
    Network Access
    Listed20 days ago

    What's inside

    Frequently Asked Questions

    This skill proactively scans your environment files to identify SDK schema changes, deprecated MCP servers, and unpinned dependencies that cause silent failures in agentic workflows. It solves the problem of 'yesterday's code' breaking due to the high release velocity of AI frameworks like Vercel AI SDK and LangGraph.

    It is compatible with any agent environment using the Model Context Protocol (MCP) or standard package managers like npm, pnpm, pip, and uv. It specifically tracks 258 packages, including core SDKs from OpenAI, Anthropic, and major open-source AI frameworks.

    No, the skill is designed to redact sensitive tokens and only transmits package names and version strings to the AgentBreakWatch index for auditing. It focuses on your manifest files like package.json and .mcp.json rather than your private application logic or environment variables.

    You receive the full audit tool capable of scanning local workspaces, access to the live AgentBreakWatch telemetry index, and pre-formatted remediation snippets for quick fixes. The purchase covers both the diagnostic scanner and the library of replacement paths for archived dependencies.

    The skill queries a live-updated telemetry index, ensuring you are protected against SDK changes that occurred as recently as 24 hours ago. This real-time data allows it to flag high-velocity 'drift' risks that static LLM knowledge or standard linters would miss.

    New in LLM Ops

    Free