Skills
Navigate
skills/sensliao/claude-code-setting/i2r-fr-authoring-mode/Gen Agent Trust Hub

i2r-fr-authoring-mode

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local Python script located at scripts/i2r.py to validate generated requirements against a schema. This is a standard part of the authoring workflow and uses a static command python scripts/i2r.py validate --stage 4.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from 03-scope.json and files in the raw/ directory. \n
  • Ingestion points: Capability clusters are identified from 03-scope.json and source references are pulled from files in raw/ (e.g., raw/idea.md).\n
  • Boundary markers: Requirements are strictly constrained to the EARS template (trigger, system_name, system_response) and must include a source_ref for traceability.\n
  • Capability inventory: The agent has the capability to run the local i2r.py script for validation.\n
  • Sanitization: A placeholder_scan is implemented to reject banned or vague terms (e.g., TBD, FIXME, quickly, securely), reducing the risk of processing ambiguous or malicious content from the source data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 09:27 AM