skills/harshanandak/forge/verify/Gen Agent Trust Hub

verify

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill extracts information from the body of a Pull Request (PR_BODY) and its branch name (PR_BRANCH) to perform actions such as closing issues. This ingestion of external content constitutes an attack surface.
  • Ingestion points: PR_BODY and PR_BRANCH are retrieved from GitHub via the gh pr view command in SKILL.md.
  • Boundary markers: The skill does not use specific delimiters or instructions to ignore embedded instructions within the parsed PR content.
  • Capability inventory: The skill utilizes git worktree remove --force for directory cleanup, git branch -d for branch deletion, and forge close to modify issue statuses.
  • Sanitization: The skill implements a validation check using forge show "$id" for every extracted ID before attempting the forge close operation, which effectively limits the impact of malicious data.
  • [COMMAND_EXECUTION]: The skill executes various shell commands using the Bash tool to manage the repository state. These include switching to master, pulling updates, listing CI workflow runs, and cleaning up worktrees. These operations are aligned with the skill's primary function of post-merge verification.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 10:07 AM
Security Audit — agent-trust-hub — verify