grace-plan

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes grace lint and grace status CLI commands. These are used for preflight validation of the change specification and to verify the project baseline before planning begins.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data from architecture specification files.
  • Ingestion points: Processes spec.xml and design-context.xml located in the .grace/ directory.
  • Boundary markers: Uses structured XML tags and a template file (references/change-plan-template.xml) to delimit content.
  • Capability inventory: Invokes grace lint and grace status; generates an XML plan containing verification commands to be processed by downstream tools.
  • Sanitization: Relies on a strict XML schema and pre-execution linting to validate inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 03:49 PM
Security Audit — agent-trust-hub — grace-plan