issue-fixer
Fail
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is vulnerable to command injection. It uses subshells and heredocs (e.g., git commit -m "$(cat <<'EOF' ... EOF)") to process untrusted issue content. Malicious input in an issue title or body containing shell expansion characters like $() or backticks would be executed by the agent's shell during command line expansion.
- [EXTERNAL_DOWNLOADS]: The skill performs bun install to fetch dependencies from public registries.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from GitHub issues.
- Ingestion points: GitHub issue titles and bodies fetched into the agent's context (SKILL.md).
- Boundary markers: No delimiters or ignore instructions are present to isolate the external data.
- Capability inventory: The agent has permissions for local filesystem modification, shell access, and remote repository push access.
- Sanitization: No validation, escaping, or filtering of the issue content is performed before it is interpolated into executable commands.
Recommendations
- AI detected serious security threats
Audit Metadata