cw-reader

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists exclusively of markdown instructions and does not contain any scripts, binaries, or executable code.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructions include reading a local reference file (../../references/context-contract.md) to establish audience context. This is a standard project-relative file access pattern for context-aware agents and does not involve hardcoded credentials or network exfiltration.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection as it ingests untrusted data from user-provided copy and project context files. However, the risk is negligible due to the following evidence chain:
  • Ingestion points: Reads ../../references/context-contract.md and user-provided text content.
  • Boundary markers: Explicit boundary markers or 'ignore' instructions are absent in the prompt templates.
  • Capability inventory: The skill has no capabilities for subprocess execution, file writing, network operations, or dynamic code evaluation.
  • Sanitization: No explicit sanitization or filtering of the input copy is defined, which is standard for text analysis tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:43 PM
Security Audit — agent-trust-hub — cw-reader