repo-fleet-standardization

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill fetches a JSON schema for validation from https://skills.sh/schemas/skills.sh.schema.json. This is a standard practice for ensuring metadata consistency and uses an official schema location for the toolset.
  • [COMMAND_EXECUTION]: The skill utilizes common developer utilities like git, gh, jq, and shellcheck to perform its tasks. These tools are appropriate for the skill's stated purpose of repository management and standardization.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external repositories (READMEs, configuration files, and GitHub metadata) and possesses capabilities to write to the filesystem and modify GitHub settings.
  • Ingestion points: Reads various files from sibling repositories, including documentation, package manifests, and CI configurations (referenced in the File-class checklist section of SKILL.md).
  • Boundary markers: The skill includes extensive 'Mutation safety' and 'Operating boundaries' sections that instruct the agent to never commit, push, or run destructive commands without explicit user approval, though it lacks technical delimiters for segregating untrusted data during processing.
  • Capability inventory: The skill can write files, delete/move files, modify repository metadata via the GitHub API, and run local validation scripts.
  • Sanitization: Uses structured parsers (jq, YAML/TOML) and linters (shellcheck) to validate modified content before finalizing changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 01:55 AM
Security Audit — agent-trust-hub — repo-fleet-standardization