roof-rfp-extract

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from external RFP and tender document PDFs, making it susceptible to indirect prompt injection.
  • Ingestion points: Content is extracted directly from user-provided roof project documents in SKILL.md.
  • Boundary markers: The instructions lack requirements for the agent to use delimiters (like XML tags) or specific instructions to ignore malicious commands embedded within the document text.
  • Capability inventory: The skill's output is written to the local file system at <rfp-dir>/roof-review-output/manifests/rfp.json, which is then consumed by downstream scripts like scripts/gate_applicability.py.
  • Sanitization: There is no evidence of content validation, escaping, or filtering for the extracted strings before they are written to the output manifest.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 12:36 AM
Security Audit — agent-trust-hub — roof-rfp-extract