github-intel

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). It includes an explicit example token ("GITHUB_TOKEN=ghp_xxx ...") and shows setting the token inline on the command line, which encourages embedding a secret verbatim into commands and would require the agent to emit the secret value in output if used.

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

• Third-party content exposure detected (high risk: 0.90). This skill's scripts (scripts/repo_analyzer.py and scripts/repo_to_markdown.py) explicitly fetch and ingest arbitrary public GitHub repo content via the GitHub API and raw.githubusercontent.com (as shown in SKILL.md and the scripts), which is untrusted user-generated content that the agent reads and uses to drive analysis and outputs, enabling indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime scripts (scripts/repo_to_markdown.py and scripts/repo_analyzer.py) make live requests to https://raw.githubusercontent.com and https://api.github.com to fetch repository files/README/tree data which are injected verbatim into the generated markdown (and thus into agent context), so external content can directly control prompts at runtime.