skills/agentic-os-org/anolisa/pr-body/Gen Agent Trust Hub

pr-body

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the diff-cover package using pip and to manage specific Rust toolchains using rustup. These downloads are performed from well-known official registries and standard tool providers.
  • [COMMAND_EXECUTION]: The skill executes local developer tools and scripts, including git, cargo, and a repository-specific Python script scripts/check-arch-boundaries.py. It also utilizes make to manage local repository configurations.
  • [PERSISTENCE]: The skill offers an optional configuration step to install git pre-push hooks using make install-hooks. This mechanism automates the execution of verification scripts during the git push process.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository's git history and file differences to generate pull request titles and descriptions.
  • Ingestion points: The skill reads content from git log and git diff within the local repository environment.
  • Boundary markers: The instructions specify natural language constraints to guide the summary process, but do not utilize technical delimiters between the data and the instructions.
  • Capability inventory: The skill can perform network operations via the GitHub CLI (gh) to create or edit PRs and issues, and can perform local file writes via build tools.
  • Sanitization: The skill relies on natural language quality rules for the AI to exclude sensitive information rather than employing automated sanitization or filtering logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:58 AM