context-engine

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill manages sensitive company information stored in a local file at ~/.claude/company-context.md. While this involves accessing sensitive data, the skill's primary purpose is to control and protect this information through a strictly defined anonymization protocol.
  • [INDIRECT_PROMPT_INJECTION]: The 'Context Enrichment' feature allows the agent to update the persistent company context file based on conversation history. This introduces a surface where malicious instructions could potentially be saved to the local profile.
  • Ingestion points: Updates are derived from in-session conversation history and new information revealed by the user (SKILL.md).
  • Boundary markers: The skill requires an explicit user confirmation ('Want me to update the file?') before modifying the context, serving as a manual validation step.
  • Capability inventory: The skill performs file read and append operations on ~/.claude/company-context.md (SKILL.md).
  • Sanitization: The skill features a detailed 'Anonymization Protocol' for outgoing data, although it does not specify sanitization procedures for data being ingested into the persistent context file.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 08:15 AM
Security Audit — agent-trust-hub — context-engine