kata-pause-work

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell utilities including grep, find, and printf to programmatically determine the current project phase and task status from local planning files.
  • [COMMAND_EXECUTION]: Executes a local Node.js script scripts/kata-lib.cjs to retrieve configuration values, specifically checking if the environment permits committing planning documents to Git.
  • [COMMAND_EXECUTION]: Performs standard Git operations (git add, git commit) to persist the handoff file, ensuring state is preserved across sessions.
  • [PROMPT_INJECTION]: Analysis identifies an indirect prompt injection surface (Category 8) where external data is incorporated into a handoff file for future consumption. 1. Ingestion points: Data is read from .planning/STATE.md and user-provided 'mental context' strings. 2. Boundary markers: The generated .continue-here.md file does not use specific delimiters to isolate external content. 3. Capability inventory: The skill can execute shell commands and local Node.js scripts. 4. Sanitization: Ingested data is written directly to the markdown template without explicit sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 04:40 PM