state-inconsistency-auditor

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions direct the agent to perform a deep analysis of untrusted source code provided by users, which creates a surface for indirect prompt injection where instructions embedded in the audited code could attempt to manipulate the agent's behavior. 1. Ingestion points: External source code files across various programming languages (Solidity, Move, Rust, etc.) provided for auditing. 2. Boundary markers: The skill methodology does not incorporate specific delimiters or 'ignore' instructions to isolate the data being analyzed from the agent's internal logic. 3. Capability inventory: The skill requires file-system read access to analyze project directories and file-system write access to output audit results to the '.audit/findings/' directory. 4. Sanitization: There is no requirement or instruction for the agent to sanitize or filter the contents of the audited files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 11:23 AM