gh-issue-to-demand-signal
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches issue titles and body snippets from public GitHub repositories and includes them directly in the agent's context for clustering and analysis.
- Ingestion points: Data is fetched from the GitHub REST API (Step 3) and processed locally (Step 4).
- Boundary markers: The skill does not use specific delimiters or explicit instructions to the agent to disregard potential instructions embedded in the external issue data.
- Capability inventory: The skill possesses file-write capabilities (to
docs/demand-signals/) and shell execution capabilities via Python scripts. - Sanitization: While the body content is truncated to 300 characters, no validation or escaping is applied to titles or snippets before they are used in prompts.
- [DYNAMIC_EXECUTION]: The skill relies on multiple Python scripts that are assembled and executed at runtime via heredocs.
- Evidence: Steps 2, 3, 4, 5, 6, 7, and 8 in
SKILL.mdgenerate and run Python code to perform API interaction and data manipulation. - Risk: These scripts interpolate user-provided repository URLs or slugs (e.g.,
raw = "REPO_INPUT_HERE"in Step 2). If the agent platform does not sanitize these inputs before substitution, it could lead to code injection into the execution environment.
Audit Metadata