brainstorm-skills

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is instructed to execute a local script, ./tests/test-codex-skills.sh, to validate global skill drafts and recommends running relevant repository tests for project-local drafts. This execution is confined to local project scripts intended for development workflows.- [PROMPT_INJECTION]: The skill processes untrusted data from the repository, creating a surface for indirect prompt injection.
  • Ingestion points: The skill reads AGENTS.md, project metadata files (such as package.json, Cargo.toml, pyproject.toml, flake.nix), and existing SKILL.md files.
  • Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore embedded instructions in the ingested files.
  • Capability inventory: The skill can write new SKILL.md files to the filesystem and execute local shell commands for testing.
  • Sanitization: There is no evidence of sanitization or filtering of the content read from the repository files before it is interpolated into the drafting process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 10:19 PM
Security Audit — agent-trust-hub — brainstorm-skills