project-context

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely composed of markdown documentation defining rules, reasons, and terminology for the repository. It does not include any scripts, executable commands, or remote dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to load specific markdown files (e.g., glossary.md, domains/layout.md) to obtain task-related context.
  • Ingestion points: glossary.md, domains/layout.md, domains/authoring.md, docs/context-engineering.md.
  • Boundary markers: None; the content is presented as authoritative project context.
  • Capability inventory: Mentions repository-specific tools such as /link-skills, /audit-project-context, and /writing-for-agents.
  • Sanitization: None; the skill relies on the integrity of the documentation within the repository environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 12:26 PM
Security Audit — agent-trust-hub — project-context