dependency-audit

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the gh-manager CLI tool to fetch dependency graphs, list Dependabot PRs, and perform pull request merges.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it processes untrusted data from external sources such as package names and pull request titles.\n
  • Ingestion points: Output from gh-manager deps graph and gh-manager deps dependabot-prs (SKILL.md) containing external package metadata.\n
  • Boundary markers: Absent; the instructions do not use delimiters to isolate external content from internal logic.\n
  • Capability inventory: The skill can perform write operations via gh-manager prs merge (SKILL.md) to modify the repository state.\n
  • Sanitization: Absent; external data is presented and processed without explicit validation or escaping. Note: The risk of automated exploitation is significantly mitigated by a mandatory human-in-the-loop approval step (AskUserQuestion) required before any merges are finalized.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 01:36 PM