skills/atman-33/agent-harness/to-prd/Gen Agent Trust Hub

to-prd

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks due to its data ingestion patterns.
  • Ingestion points: Processes conversation history and codebase context to generate the PRD.
  • Boundary markers: Absent. There are no instructions to disregard embedded commands or delimiters within the ingested data.
  • Capability inventory: The skill possesses the ability to browse the repository and write output to an external project issue tracker.
  • Sanitization: Absent. The skill does not implement validation or escaping for the data it synthesizes before publishing it.
  • [COMMAND_EXECUTION]: The instructions reference a setup command /setup-matt-pocock-skills which is intended to initialize the project environment and metadata vocabulary.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 12:04 PM
Security Audit — agent-trust-hub — to-prd