deliverable-fidelity-pl
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to execute a local Node.js script (scripts/fidelity-check.mjs). This script performs text normalization and keyword matching to verify document content. The script is self-contained and does not use external dependencies or network modules. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted text from external documents (findings and deliverables).
- Ingestion points: Data enters through the
task.findingsandtask.deliverablefields in a JSON file passed to the script. - Boundary markers: The instructions lack explicit boundary markers or instructions to ignore embedded commands within the deliverable text during the LLM spot-check phase.
- Capability inventory: The skill is restricted to local bash execution and file reading; it has no network access or file-writing capabilities.
- Sanitization: The Node.js script performs basic text normalization (lowercase conversion and whitespace stripping) but does not implement security sanitization for LLM contexts.
Audit Metadata