idea-screener
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from
IDEA_DIRECTION_POOL.jsonl, which is a vulnerability surface for indirect prompt injection. Maliciously crafted content within these files could influence the agent's behavior during the scoring process or affect downstream tasks that consume the generated markdown and JSONL outputs. - Ingestion points: The script
scripts/run.pyreads data fromworkspace / 'output' / 'trace' / 'IDEA_DIRECTION_POOL.jsonl'. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands were found in the processing logic.
- Capability inventory: The script performs both file read and file write operations (
read_jsonl,write_markdown,write_jsonl). - Sanitization: The implementation does not demonstrate explicit sanitization or validation of the text content within the
DirectionCardobjects before they are passed to the scoring and rendering functions.
Audit Metadata