lov-gh-tidy
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) andgitto manage repository resources. It performs read operations to scan the repository state and write operations to close issues, merge pull requests, and delete branches or labels. These actions are within the stated scope of the skill and include a user confirmation step. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external GitHub repositories.
- Ingestion points:
SKILL.md(Step 1: Scan) usesgh issue list,gh pr list, andgh label listto fetch repository content. - Boundary markers: Absent. The instructions do not define clear delimiters or warnings to the agent regarding embedded instructions in the fetched data.
- Capability inventory:
SKILL.md(Step 4: Execute) includes capabilities to modify the repository, such as closing issues, merging pull requests, and deleting remote branches and labels. - Sanitization: Absent. There is no evidence of filtering or validation of the data retrieved from GitHub before it is summarized and presented for triage.
Audit Metadata