spec-driven-archive

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes Node.js to run a local script spec-driven.js for listing, checking, and archiving changes. This is standard operational behavior for this type of development utility.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes markdown files from the .spec-driven directory which may contain untrusted content.
  • Ingestion points: Files located in .spec-driven/changes/<name>/specs/ are read and merged into main specification files.
  • Boundary markers: The process relies on ### Requirement: headers; however, there are no specific instructions or delimiters provided to prevent the agent from interpreting embedded instructions within these requirement blocks as commands.
  • Capability inventory: The agent has capabilities to read/write files in the local repository and execute Node.js commands based on the context of these specifications.
  • Sanitization: The skill does not perform sanitization or validation of the requirement text before merging or summarizing changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 07:34 AM