proposal-review

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and assess user-provided thesis proposals (markdown files), which creates an attack surface for indirect prompt injection. A maliciously crafted proposal could contain instructions that attempt to influence the agent's behavior during the review process.
  • Ingestion points: The agent ingests user-supplied markdown files from the workspace as specified in the SKILL.md instructions and evals/evals.json fixtures.
  • Boundary markers: The instructions do not specify the use of delimiters or instructions for the agent to ignore or isolate commands that may be embedded within the proposal content.
  • Capability inventory: The skill reads the proposal file, accesses local reference files (references/guidelines.md and structure.json), and writes an output review file (<slug>-review.md) to the local directory.
  • Sanitization: There is no evidence of content sanitization or validation performed on the ingested proposal text to filter out potential instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:46 AM
Security Audit — agent-trust-hub — proposal-review