team-fix
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user arguments or external GitHub issues without explicit boundary markers or sanitization.
- Ingestion points:
references/01-input.md(processing$ARGUMENTSandgh issue viewoutput). - Boundary markers: Absent; no specific delimiters or instructions to ignore embedded commands in the bug reports are specified in the provided instructions.
- Capability inventory: Git operations (
git commit,git push), GitHub CLI operations (gh pr create), file system writes, and subagent dispatch (references/06-execution.md). - Sanitization: Absent; the skill does not explicitly verify or sanitize the content of bug descriptions or issue titles/bodies before processing or creating task documentation.
- [COMMAND_EXECUTION]: The skill executes various shell commands and internal scripts for pipeline management.
- Executes
scripts/branch-gate.shandscripts/plan-id.sh(a referenced internal script not provided in the source) to manage branching and task identification. - Uses standard development tools including
git(fetch, worktree, switch, commit, push) andgh(issue view, pr view, pr create). - [EXTERNAL_DOWNLOADS]: The skill fetches data from external developer services.
- Uses
gh issue viewto retrieve issue content from GitHub. - Uses
git fetchto update local repository state from remote origins. - These operations target established, well-known platforms used for software development.
Audit Metadata