add-linear

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly prompts for OAuth client ID/secret and webhook signing secret and then injects those answers into an nc:env-set directive ({{...}}), which requires the agent to output the secret values verbatim into the generated .env content.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow is webhook-driven for Linear issue comment threads (a Linear team webhook “Comment” event delivers free-text comments to the agent), so outsider-authored comment bodies are ingested at runtime via the Linear adapter.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill installs and imports the external package @chat-adapter/linear@4.29.0 at runtime (installed in SKILL.md and exercised by the registration test), which fetches and executes third-party adapter code inside the agent process.