implement-next-phase

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill automatically executes git add -A at the beginning of each work phase to checkpoint current changes.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied 'sketch plan' files to guide implementation, which presents a surface for indirect prompt injection.
  • Ingestion points: SKILL.md (First actions Step 1 parses the sketch plan file and Phase N heading).
  • Boundary markers: The skill lacks explicit delimiters to isolate external plan content or instructions to ignore embedded commands within those files.
  • Capability inventory: The skill has permissions to read/write local files and execute git commands.
  • Sanitization: No specific sanitization or schema validation is mentioned for the contents of the external plan files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 06:58 PM
Security Audit — agent-trust-hub — implement-next-phase