detect-duplicates
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests and processes user-generated feature request data (titles and descriptions) from an external API. This content is used to inform the agent's logic for duplicate detection and insight generation, creating a surface for indirect prompt injection.
- Ingestion points: Data is fetched via
wccom-feature-requests-listin Step 2. - Boundary markers: None explicitly defined to isolate untrusted user text from agent instructions.
- Capability inventory: The agent can update request statuses, post comments to the marketplace, and execute scripts.
- Sanitization: No explicit sanitization or filtering of the feature request text is mentioned before processing.
- [COMMAND_EXECUTION]: The skill invokes a local Python script (
scripts/detect_duplicates.py) using shell commands to process JSONL data files. This involves interpolating file paths and identifiers into command arguments. - Evidence:
python3 scripts/detect_duplicates.py $TMPDIR/frs-<product_id>.jsonl --jsonfound in Step 3.
Audit Metadata