Skip to content

aiskillstore/marketplace / pending/getrunheed/runheed

runheed

Verified

Error tracking your coding agent can read. Use when the user wants production errors from a Node.js or Python app delivered to Claude Code, Cursor, Codex or any MCP client so the agent can read the stack trace, fix it and mark it resolved. Also use when the user keeps pasting stack traces into chat, asks "what broke in production", or wants an error tracker with an MCP server.

FileKindSize
SKILL.mdskill md6.3 KB
skill-report.jsonconfig45.8 KB

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-11-18T09:17:57.761Z",
    "slug": "getrunheed-runheed",
    "source_url": "https://github.com/getrunheed/runheed-plugin/tree/d70b2e63846bad4c6912a958cace20745e15ddf1/skills/runheed",
    "source_ref": "d70b2e63846bad4c6912a958cace20745e15ddf1",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "6932209f543740fd06c69abb79bf5e37cd56af43e3c2cb20e14daaae462fcb94",
    "tree_hash": "9ddb3d5071e3ba3d9a9793896015e3df4d00aeea2e3231713932d1215e3c07a7",
    "provenance": {
      "model_effective": "codex",
      "fallback_chain": [
        {
          "agent": "codex",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    },
    "upstream_version_raw": "1.0.0",
    "upstream_version_normalized": "1.0.0",
    "upstream_version_source": "version",
    "upstream_version_status": "valid",
    "upstream_commit_sha": "d70b2e63846bad4c6912a958cace20745e15ddf1"
  },
  "skill": {
    "name": "runheed",
    "description": "Connect Runheed error tracking to a compatible MCP client. Review production errors from Node.js and Python applications, investigate stack traces, and mark errors resolved after deployment.",
    "summary": "Review production errors and track their resolution from a compatible MCP client.",
    "icon": "📦",
    "version": "1.0.0",
    "author": "getrunheed",
    "category": "devops",
    "tags": [
      "error tracking",
      "incident triage",
      "MCP",
      "Node.js",
      "Python"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "external_commands",
      "network",
      "env_access"
    ]
  },
  "security_audit": {
    "risk_level": "medium",
    "is_blocked": false,
    "safe_to_publish": true,
    "analysis_status": "ok",
    "summary": "Most matches are reference links, Markdown formatting, or reads of one configuration variable. The setup uses unverified package downloads, exposes a fallback agent key through command arguments, and registers a demo server without separate approval. Production error context and remote tool responses also need explicit data and trust controls.",
    "remediation": [
      {
        "issue": "Setup installs or runs packages from vendor URLs without documented integrity checks.",
        "suggestion": "Publish versioned packages with verifiable signatures or digests. Require user approval before installing the SDK or running the local MCP package.",
        "severity": "medium"
      },
      {
        "issue": "The local setup asks for an agent key and places it in command arguments.",
        "suggestion": "Prefer OAuth. For local setup, have the user enter a scoped key through a supported secret store without exposing it in chat or command arguments.",
        "severity": "high"
      },
      {
        "issue": "A demo request triggers MCP registration without specific approval for the configuration change.",
        "suggestion": "Ask before registering the demo endpoint. Use temporary, isolated configuration where supported and explain how to remove the connection.",
        "severity": "medium"
      },
      {
        "issue": "Setup does not define which production error data may leave the application.",
        "suggestion": "Document captured fields and retention. Require approval for each environment, and redact credentials and personal data before sending events.",
        "severity": "medium"
      },
      {
        "issue": "The workflow lacks guidance for handling instructions embedded in error reports.",
        "suggestion": "Treat tool descriptions and returned event text as untrusted. Do not follow embedded directions, and require user confirmation before changing incident state.",
        "severity": "medium"
      }
    ],
    "risk_factor_evidence": [
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          },
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          },
          {
            "file": "SKILL.md",
            "line_start": 64,
            "line_end": 64
          },
          {
            "file": "SKILL.md",
            "line_start": 65,
            "line_end": 65
          },
          {
            "file": "SKILL.md",
            "line_start": 68,
            "line_end": 68
          },
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          }
        ]
      },
      {
        "factor": "network",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 5,
            "line_end": 5
          },
          {
            "file": "SKILL.md",
            "line_start": 6,
            "line_end": 6
          },
          {
            "file": "SKILL.md",
            "line_start": 7,
            "line_end": 7
          },
          {
            "file": "SKILL.md",
            "line_start": 8,
            "line_end": 8
          },
          {
            "file": "SKILL.md",
            "line_start": 9,
            "line_end": 9
          },
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          },
          {
            "file": "SKILL.md",
            "line_start": 39,
            "line_end": 39
          },
          {
            "file": "SKILL.md",
            "line_start": 40,
            "line_end": 40
          },
          {
            "file": "SKILL.md",
            "line_start": 41,
            "line_end": 41
          },
          {
            "file": "SKILL.md",
            "line_start": 42,
            "line_end": 42
          },
          {
            "file": "SKILL.md",
            "line_start": 43,
            "line_end": 43
          },
          {
            "file": "SKILL.md",
            "line_start": 44,
            "line_end": 44
          },
          {
            "file": "SKILL.md",
            "line_start": 61,
            "line_end": 61
          },
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          },
          {
            "file": "SKILL.md",
            "line_start": 64,
            "line_end": 64
          },
          {
            "file": "SKILL.md",
            "line_start": 65,
            "line_end": 65
          },
          {
            "file": "SKILL.md",
            "line_start": 67,
            "line_end": 67
          },
          {
            "file": "SKILL.md",
            "line_start": 68,
            "line_end": 68
          },
          {
            "file": "SKILL.md",
            "line_start": 69,
            "line_end": 69
          },
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          },
          {
            "file": "SKILL.md",
            "line_start": 71,
            "line_end": 71
          },
          {
            "file": "SKILL.md",
            "line_start": 74,
            "line_end": 74
          }
        ]
      },
      {
        "factor": "env_access",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          },
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          },
          {
            "file": "SKILL.md",
            "line_start": 64,
            "line_end": 64
          },
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "external_commands:SKILL.md:35:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 35,
        "line_end": 35,
        "snippet": "If the user wants to see it work before paying, connect yourself to the demo server first: it holds "
      },
      {
        "id": "external_commands:SKILL.md:63:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ "
      },
      {
        "id": "external_commands:SKILL.md:64:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 64,
        "line_end": 64,
        "snippet": "- Python: `pip install https://runheed.com/dl/runheed-0.1.5-py3-none-any.whl` then `runheed.init(dsn"
      },
      {
        "id": "external_commands:SKILL.md:65:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 65,
        "line_end": 65,
        "snippet": "- Serverless (Vercel, Lambda, Netlify, Cloudflare) and Next.js: the Node.js SDK batches events for u"
      },
      {
        "id": "external_commands:SKILL.md:68:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 68,
        "line_end": 68,
        "snippet": "`claude mcp add --transport http runheed https://runheed.com/api/mcp`, then /mcp > runheed > Authent"
      },
      {
        "id": "external_commands:SKILL.md:70:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 70,
        "line_end": 70,
        "snippet": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0."
      },
      {
        "id": "network:SKILL.md:5:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 5,
        "line_end": 5,
        "snippet": "homepage: https://runheed.com"
      },
      {
        "id": "network:SKILL.md:6:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 6,
        "line_end": 6,
        "snippet": "pricing: https://runheed.com/pricing"
      },
      {
        "id": "network:SKILL.md:7:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 7,
        "line_end": 7,
        "snippet": "llms: https://runheed.com/llms.txt"
      },
      {
        "id": "network:SKILL.md:8:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 8,
        "line_end": 8,
        "snippet": "setup: https://runheed.com/SETUP.txt"
      },
      {
        "id": "network:SKILL.md:9:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 9,
        "line_end": 9,
        "snippet": "checkout: https://runheed.com/dashboard/start?plan={solo|pro|scale}&interval={month|year}&code=RUNHE"
      },
      {
        "id": "network:SKILL.md:35:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 35,
        "line_end": 35,
        "snippet": "If the user wants to see it work before paying, connect yourself to the demo server first: it holds "
      },
      {
        "id": "network:SKILL.md:39:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 39,
        "line_end": 39,
        "snippet": "- Solo monthly: https://runheed.com/dashboard/start?plan=solo&interval=month&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:40:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 40,
        "line_end": 40,
        "snippet": "- Solo yearly: https://runheed.com/dashboard/start?plan=solo&interval=year&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:41:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 41,
        "line_end": 41,
        "snippet": "- Pro monthly: https://runheed.com/dashboard/start?plan=pro&interval=month&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:42:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 42,
        "line_end": 42,
        "snippet": "- Pro yearly: https://runheed.com/dashboard/start?plan=pro&interval=year&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:43:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 43,
        "line_end": 43,
        "snippet": "- Scale monthly: https://runheed.com/dashboard/start?plan=scale&interval=month&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:44:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 44,
        "line_end": 44,
        "snippet": "- Scale yearly: https://runheed.com/dashboard/start?plan=scale&interval=year&code=RUNHEED-AGENT"
      },
      {
        "id": "network:SKILL.md:61:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 61,
        "line_end": 61,
        "snippet": "1. Ask the user to create a project at https://runheed.com/dashboard/projects and give you its DSN ("
      },
      {
        "id": "network:SKILL.md:63:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ "
      },
      {
        "id": "network:SKILL.md:64:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 64,
        "line_end": 64,
        "snippet": "- Python: `pip install https://runheed.com/dl/runheed-0.1.5-py3-none-any.whl` then `runheed.init(dsn"
      },
      {
        "id": "network:SKILL.md:65:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 65,
        "line_end": 65,
        "snippet": "- Serverless (Vercel, Lambda, Netlify, Cloudflare) and Next.js: the Node.js SDK batches events for u"
      },
      {
        "id": "network:SKILL.md:67:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 67,
        "line_end": 67,
        "snippet": "3. Add the remote MCP server (https://runheed.com/api/mcp, OAuth sign-in, no key to copy):"
      },
      {
        "id": "network:SKILL.md:68:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 68,
        "line_end": 68,
        "snippet": "`claude mcp add --transport http runheed https://runheed.com/api/mcp`, then /mcp > runheed > Authent"
      },
      {
        "id": "network:SKILL.md:69:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 69,
        "line_end": 69,
        "snippet": "Without remote support: ask the user for an agent key from https://runheed.com/dashboard/settings (s"
      },
      {
        "id": "network:SKILL.md:70:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 70,
        "line_end": 70,
        "snippet": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0."
      },
      {
        "id": "network:SKILL.md:71:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 71,
        "line_end": 71,
        "snippet": "Other clients: https://runheed.com/connect"
      },
      {
        "id": "network:SKILL.md:74:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 74,
        "line_end": 74,
        "snippet": "Full step-by-step instructions: https://runheed.com/SETUP.txt"
      },
      {
        "id": "env_access:SKILL.md:63:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "SKILL.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ "
      },
      {
        "id": "env_access:SKILL.md:63:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "SKILL.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ "
      },
      {
        "id": "env_access:SKILL.md:64:python-environment-access",
        "category": "env_access",
        "severity": "low",
        "pattern": "Python environment access",
        "file": "SKILL.md",
        "line_start": 64,
        "line_end": 64,
        "snippet": "- Python: `pip install https://runheed.com/dl/runheed-0.1.5-py3-none-any.whl` then `runheed.init(dsn"
      },
      {
        "id": "env_access:SKILL.md:70:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "SKILL.md",
        "line_start": 70,
        "line_end": 70,
        "snippet": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0."
      },
      {
        "id": "sensitive:SKILL.md:63:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "SKILL.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ "
      },
      {
        "id": "blocker:SKILL.md:18:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "SKILL.md",
        "line_start": 18,
        "line_end": 18,
        "snippet": "- Paid only, flat price per workspace, no seats, no overage bills (events over the limit are rejecte"
      }
    ],
    "finding_verdicts": [
      {
        "id": "external_commands:SKILL.md:35:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The backticks format text, but the instruction tells the agent to run a CLI command that adds a demo MCP server to its configuration. It does not ask separately before making that change."
      },
      {
        "id": "external_commands:SKILL.md:63:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The Markdown shows an actual npm install step, not shell substitution. Installing the unverified tarball can execute package lifecycle scripts and changes the application dependencies."
      },
      {
        "id": "external_commands:SKILL.md:64:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The Markdown shows an actual pip install step for a vendor-hosted wheel. Application initialization then imports and runs that installed code without a documented integrity check."
      },
      {
        "id": "external_commands:SKILL.md:65:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The backticks delimit a JavaScript SDK call for flushing pending reports. They do not invoke a shell or another program."
      },
      {
        "id": "external_commands:SKILL.md:68:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.91,
        "reason": "The command registers the expected Runheed endpoint during paid setup, not a local executable or a shell substitution. The user must authorize access through OAuth."
      },
      {
        "id": "external_commands:SKILL.md:70:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The fallback configures a local MCP process that runs downloaded package code. The execution command skips the package manager prompt and supplies an agent credential to that process."
      },
      {
        "id": "network:SKILL.md:5:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "This homepage URL is descriptive metadata. It does not instruct the agent to connect or transfer data."
      },
      {
        "id": "network:SKILL.md:6:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "This metadata points to a public pricing page. It does not initiate checkout or send workspace data."
      },
      {
        "id": "network:SKILL.md:7:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "This is a metadata reference to vendor documentation. The skill does not instruct the agent to fetch or follow it automatically."
      },
      {
        "id": "network:SKILL.md:8:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This metadata points to setup documentation. It is not an instruction to fetch or execute anything."
      },
      {
        "id": "network:SKILL.md:9:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The checkout template carries plan, billing interval, and discount parameters. The workflow leaves opening the link and paying to the user."
      },
      {
        "id": "network:SKILL.md:35:hardcoded-url",
        "verdict": "confirmed",
        "confidence": 0.98,
        "reason": "The URL is a remote MCP endpoint the skill tells the agent to use for a demo. Its outputs enter the agent session without a separate approval step for that connection."
      },
      {
        "id": "network:SKILL.md:39:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Solo monthly link passes a plan choice and discount to browser checkout. The user opens it and controls payment."
      },
      {
        "id": "network:SKILL.md:40:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Solo yearly link is for user-initiated checkout. Its parameters select terms and do not contain credentials."
      },
      {
        "id": "network:SKILL.md:41:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Pro monthly link preselects a plan for browser checkout. It does not authorize an agent purchase."
      },
      {
        "id": "network:SKILL.md:42:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Pro yearly URL is a checkout link intended for the user. It contains plan and discount parameters, not private workspace data."
      },
      {
        "id": "network:SKILL.md:43:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Scale monthly link is presented for the user to open. It does not charge the account through an agent action."
      },
      {
        "id": "network:SKILL.md:44:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The Scale yearly link selects a plan and interval before browser checkout. It leaves the payment decision to the user."
      },
      {
        "id": "network:SKILL.md:61:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "The dashboard URL is for the user to create the intended project. The adjacent DSN is not appended to that dashboard URL or transmitted by clicking it."
      },
      {
        "id": "network:SKILL.md:63:hardcoded-url",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The URL is an install source for the Node.js SDK. The versioned filename does not verify the archive contents, and the instructions provide no digest or signature."
      },
      {
        "id": "network:SKILL.md:64:hardcoded-url",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The first URL is used to download and install a Python wheel without documented integrity verification. The other URLs are framework documentation."
      },
      {
        "id": "network:SKILL.md:65:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The URL links to Next.js documentation. The adjacent SDK flush call does not use that link as its network destination."
      },
      {
        "id": "network:SKILL.md:67:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL identifies the disclosed service endpoint for the requested integration. The setup requires user OAuth consent and does not put a key in the URL."
      },
      {
        "id": "network:SKILL.md:68:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The CLI registers the same declared service endpoint. The instructions require the user to sign in and approve access before private errors can be retrieved."
      },
      {
        "id": "network:SKILL.md:69:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "The settings link points to the vendor dashboard where the user obtains a key. The URL itself carries no key and does not send one elsewhere."
      },
      {
        "id": "network:SKILL.md:70:hardcoded-url",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The URL supplies the archive for the local MCP executable. The instructions run it without verifying the downloaded bytes."
      },
      {
        "id": "network:SKILL.md:71:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This is a setup reference for other MCP clients. The line does not ask the agent to send credentials or execute downloaded code."
      },
      {
        "id": "network:SKILL.md:74:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This is a reference to vendor setup documentation. The link itself does not initiate an install or send data."
      },
      {
        "id": "env_access:SKILL.md:63:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The Node.js example reads only the designated Runheed DSN for SDK initialization. It does not enumerate or print unrelated environment values."
      },
      {
        "id": "env_access:SKILL.md:63:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The environment object is used to select one named configuration value. The whole object is not passed to the SDK or logged."
      },
      {
        "id": "env_access:SKILL.md:64:python-environment-access",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The Python example looks up only the Runheed DSN and passes it to SDK initialization. It does not dump the environment."
      },
      {
        "id": "env_access:SKILL.md:70:generic-api-secret-keys",
        "verdict": "confirmed",
        "confidence": 0.99,
        "reason": "The shown value is a placeholder, but setup tells the agent to substitute a real key into CLI arguments. That can expose it through command history, execution logs, or process inspection."
      },
      {
        "id": "sensitive:SKILL.md:63:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The Node.js snippet reads one process environment variable. No evidence found of reading or writing an environment file on this line."
      },
      {
        "id": "blocker:SKILL.md:18:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The line describes billing and event limits. It does not request host discovery, file inspection, or system enumeration."
      }
    ],
    "semantic_findings": [
      {
        "title": "Production Error Data Shared Without Redaction Guidance",
        "description": "The documented setup reports production failures to a hosted service and exposes stack traces, breadcrumbs, and session context to the agent. These fields may contain sensitive details.",
        "severity": "medium",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 14,
            "line_end": 14
          },
          {
            "file": "SKILL.md",
            "line_start": 61,
            "line_end": 66
          }
        ],
        "confidence": 0.95,
        "confidence_reasoning": "The skill describes automatic reporting and rich error context but gives no pre-ingestion redaction or review steps. The actual SDK filters were not verified."
      },
      {
        "title": "Untrusted Error Text Can Influence Agent Actions",
        "description": "The workflow uses remote error details to guide code changes and incident resolution. It does not instruct the agent to treat error content as untrusted.",
        "severity": "medium",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 14,
            "line_end": 14
          },
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          },
          {
            "file": "SKILL.md",
            "line_start": 67,
            "line_end": 72
          }
        ],
        "confidence": 0.9,
        "confidence_reasoning": "Error text can carry user-controlled strings, and the documented workflow lets returned results guide decisions. No evidence found of an actual injection payload in the skill."
      }
    ],
    "critical_findings": [],
    "high_findings": [
      {
        "title": "Generic API/secret keys",
        "description": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          }
        ],
        "review_kind": "capability",
        "source_category": "env_access",
        "source_severity": "high",
        "confidence": 0.99,
        "confidence_reasoning": "The shown value is a placeholder, but setup tells the agent to substitute a real key into CLI arguments. That can expose it through command history, execution logs, or process inspection."
      }
    ],
    "medium_findings": [
      {
        "title": "Ruby/shell backtick execution",
        "description": "If the user wants to see it work before paying, connect yourself to the demo server first: it holds ",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.99,
        "confidence_reasoning": "The backticks format text, but the instruction tells the agent to run a CLI command that adds a demo MCP server to its configuration. It does not ask separately before making that change."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ ",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.99,
        "confidence_reasoning": "The Markdown shows an actual npm install step, not shell substitution. Installing the unverified tarball can execute package lifecycle scripts and changes the application dependencies."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "- Python: `pip install https://runheed.com/dl/runheed-0.1.5-py3-none-any.whl` then `runheed.init(dsn",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 64,
            "line_end": 64
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.99,
        "confidence_reasoning": "The Markdown shows an actual pip install step for a vendor-hosted wheel. Application initialization then imports and runs that installed code without a documented integrity check."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.99,
        "confidence_reasoning": "The fallback configures a local MCP process that runs downloaded package code. The execution command skips the package manager prompt and supplies an agent credential to that process."
      },
      {
        "title": "Production Error Data Shared Without Redaction Guidance",
        "description": "The documented setup reports production failures to a hosted service and exposes stack traces, breadcrumbs, and session context to the agent. These fields may contain sensitive details.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 14,
            "line_end": 14
          },
          {
            "file": "SKILL.md",
            "line_start": 61,
            "line_end": 66
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.95,
        "confidence_reasoning": "The skill describes automatic reporting and rich error context but gives no pre-ingestion redaction or review steps. The actual SDK filters were not verified."
      },
      {
        "title": "Untrusted Error Text Can Influence Agent Actions",
        "description": "The workflow uses remote error details to guide code changes and incident resolution. It does not instruct the agent to treat error content as untrusted.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 14,
            "line_end": 14
          },
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          },
          {
            "file": "SKILL.md",
            "line_start": 67,
            "line_end": 72
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.9,
        "confidence_reasoning": "Error text can carry user-controlled strings, and the documented workflow lets returned results guide decisions. No evidence found of an actual injection payload in the skill."
      }
    ],
    "low_findings": [
      {
        "title": "Hardcoded URL",
        "description": "If the user wants to see it work before paying, connect yourself to the demo server first: it holds ",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 35,
            "line_end": 35
          }
        ],
        "review_kind": "capability",
        "source_category": "network",
        "source_severity": "low",
        "confidence": 0.98,
        "confidence_reasoning": "The URL is a remote MCP endpoint the skill tells the agent to use for a demo. Its outputs enter the agent session without a separate approval step for that connection."
      },
      {
        "title": "Hardcoded URL",
        "description": "- Node.js / TypeScript: `npm install https://runheed.com/dl/runheed-0.1.5.tgz` then `runheed.init({ ",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 63,
            "line_end": 63
          }
        ],
        "review_kind": "capability",
        "source_category": "network",
        "source_severity": "low",
        "confidence": 0.99,
        "confidence_reasoning": "The URL is an install source for the Node.js SDK. The versioned filename does not verify the archive contents, and the instructions provide no digest or signature."
      },
      {
        "title": "Hardcoded URL",
        "description": "- Python: `pip install https://runheed.com/dl/runheed-0.1.5-py3-none-any.whl` then `runheed.init(dsn",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 64,
            "line_end": 64
          }
        ],
        "review_kind": "capability",
        "source_category": "network",
        "source_severity": "low",
        "confidence": 0.99,
        "confidence_reasoning": "The first URL is used to download and install a Python wheel without documented integrity verification. The other URLs are framework documentation."
      },
      {
        "title": "Hardcoded URL",
        "description": "`claude mcp add runheed --env RUNHEED_API_KEY=al_... -- npx -y https://runheed.com/dl/runheed-mcp-0.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 70,
            "line_end": 70
          }
        ],
        "review_kind": "capability",
        "source_category": "network",
        "source_severity": "low",
        "confidence": 0.99,
        "confidence_reasoning": "The URL supplies the archive for the local MCP executable. The instructions run it without verifying the downloaded bytes."
      }
    ],
    "dangerous_patterns": [],
    "files_scanned": 1,
    "total_lines": 81,
    "audit_model": "codex",
    "audited_at": "2026-11-18T09:17:57.761Z",
    "agent_auto_install_policy": "allowed",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Investigate Production Errors with Runheed",
    "value_statement": "Pasted stack traces often lack the context needed to diagnose a production error. Runheed lets a compatible MCP client review error occurrences and track their resolution.",
    "seo_keywords": [
      "Claude",
      "Codex",
      "Claude Code",
      "Runheed",
      "production error tracking",
      "MCP error monitoring",
      "Node.js error monitoring",
      "Python error monitoring",
      "stack trace analysis"
    ],
    "actual_capabilities": [
      "Retrieve recent errors from a connected Runheed workspace.",
      "Inspect individual occurrences, stack traces, breadcrumbs, and available session context.",
      "Review synthetic errors through the public demo without a paid workspace.",
      "Guide SDK setup for Node.js and Python and verify ingestion with a test event.",
      "Mark a Runheed error resolved after the fix ships."
    ],
    "limitations": [
      "Private project monitoring requires a paid Runheed workspace.",
      "Automatic capture uses first-party Node.js and Python SDKs, not browser or mobile SDKs.",
      "Runheed does not provide performance tracing, session replay, or release health.",
      "The skill does not automatically create issues in GitHub, Linear, or Jira."
    ],
    "use_cases": [
      {
        "title": "Review a Sample Incident",
        "description": "Use an approved demo connection to see how stack traces and breadcrumbs appear before connecting a real application.",
        "target_user": "Independent application developer"
      },
      {
        "title": "Check Staging Error Capture",
        "description": "Verify that a planned test event from a staging application appears in the expected Runheed project.",
        "target_user": "Platform engineer"
      },
      {
        "title": "Triage a Production Error",
        "description": "Review a recent error and compare occurrences to identify what to investigate before proposing a fix.",
        "target_user": "On-call engineer"
      }
    ],
    "prompt_templates": [
      {
        "title": "Explain a Demo Error",
        "prompt": "Use my existing Runheed demo connection. Summarize one sample error and explain the stack trace in plain language.",
        "scenario": "Review a sample without connecting a production project."
      },
      {
        "title": "Summarize Recent Errors",
        "prompt": "List recent Runheed errors for my selected project. Give each error a short summary and omit sensitive values.",
        "scenario": "Get a read-only view of recent project failures."
      },
      {
        "title": "Compare Error Occurrences",
        "prompt": "Compare the stack traces and breadcrumbs for this error. Identify the recurring failure point, separate observations from hypotheses, and suggest a regression test.",
        "scenario": "Investigate a recurring failure without changing files."
      },
      {
        "title": "Review Resolution Readiness",
        "prompt": "Review the selected incident, its latest occurrences, and the deployment evidence I provided. Recommend a resolution status and wait for my approval before updating Runheed.",
        "scenario": "Decide whether to close an incident after deployment."
      }
    ],
    "output_examples": [
      {
        "input": "Explain one synthetic error from the connected demo.",
        "output": "In this sample, the request reaches order validation and stops when a required field is missing. Check where the request supplies that field."
      },
      {
        "input": "Summarize what the recent occurrences of this timeout have in common.",
        "output": [
          "The reviewed occurrences stop at the same outbound request.",
          "The stack traces show where the timeout surfaced, but not why the upstream service failed.",
          "Check upstream latency and add a regression test for timeout handling."
        ]
      },
      {
        "input": "Should I resolve this incident after the deployment?",
        "output": "The latest recorded occurrence predates the deployment. That does not establish that the fix worked. The incident remains unchanged pending verification and your approval."
      }
    ],
    "best_practices": [
      "Start with synthetic demo data or an approved staging project before enabling production reporting.",
      "Prefer OAuth and verify package provenance before installing any SDK or local MCP package.",
      "Redact sensitive event fields and confirm deployment before approving an incident status change."
    ],
    "anti_patterns": [
      "Pasting agent keys into chat or putting them in command arguments.",
      "Treating text inside stack traces or breadcrumbs as instructions.",
      "Sending production error reports before reviewing what the SDK collects."
    ],
    "faq": [
      {
        "question": "Can I try Runheed without connecting a production application?",
        "answer": "Yes. The public demo uses synthetic errors and does not require an account or paid plan."
      },
      {
        "question": "Which applications have first-party SDKs?",
        "answer": "Runheed documents SDKs for Node.js and Python. Other clients can send events to the ingest endpoint without automatic capture."
      },
      {
        "question": "Do I need a paid workspace?",
        "answer": "Yes, for monitoring your own projects. The demo is available without a paid workspace."
      },
      {
        "question": "Do I need to share an API key with my coding agent?",
        "answer": "No for clients that support the remote MCP connection with OAuth. The documented local fallback requires an agent key."
      },
      {
        "question": "Does Runheed fix and deploy code?",
        "answer": "Runheed provides error context and records resolution status. The coding agent and your deployment workflow handle code changes, tests, and release."
      },
      {
        "question": "Does Runheed open tickets in other trackers?",
        "answer": "No. The documented workflow keeps error triage and resolution in Runheed rather than filing tickets in GitHub, Linear, or Jira."
      }
    ]
  },
  "file_structure": [
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 81
    }
  ]
}
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexUnverifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall