agentic-actions-auditor

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill requires fetching and reporting workflow YAML (including evidence snippets) and explicitly captures fields like token and env var values, which forces the agent to read and potentially emit embedded secret values verbatim in its findings.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly performs runtime GitHub API fetches of workflow YAML (e.g., "gh api repos/{owner}/{repo}/contents/.github/workflows" and "gh api repos/{owner}/{repo}/contents/.github/workflows/{filename}?ref={ref}") and those fetched files are meant to be read and injected into the agent's analysis context to extract prompt fields, so remote content can directly influence the agent's prompts at runtime.