triage-verifier
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted vulnerability research data (claims, reproduction steps, and impact descriptions) and interpolate it into markdown proof packets. While this is an ingestion surface for potentially malicious payloads, it is the primary intended purpose of the skill.
- Ingestion points: The
scripts/proof_packet.pyscript accepts finding metadata as command-line arguments. - Boundary markers: The generated markdown template uses standard headers (
#,##) and bullet points to separate data fields. - Capability inventory: The script performs file-write operations to save reports in the
data/triage-verifier/directory. - Sanitization: The script includes a
slug()function that sanitizes the finding ID for use in the filename using a regular expression ([^a-zA-Z0-9_.-]+). The content of the report is interpolated as raw strings into the template.
Audit Metadata