context-apply
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from local files that could influence agent behavior.
- Ingestion points: The agent is directed to read
.context/INDEX.mdand linked files in the.context/directory (SKILL.md). - Boundary markers: No specific boundary markers or "ignore embedded instructions" warnings are defined to protect the agent from malicious content or instructions hidden within knowledge entries.
- Capability inventory: The skill is intended for use during planning, implementation, and validation, where the agent likely has access to file-system modification and command execution tools.
- Sanitization: There is no instruction to sanitize, validate, or escape content retrieved from the knowledge entries before transforming rules into actions or checks.
Audit Metadata