workflow-implement-spec

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon untrusted external data, specifically "canonical specs" and user-provided prompts, which may contain malicious instructions targeting the agent's automated workflow.
  • Ingestion points: Ingests external specification files (referenced as spec_baseline) and user-provided prompts to determine implementation tasks in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the specification data are present.
  • Capability inventory: The workflow has the authority to modify files (via $plan --mode=execute), spawn subagents (spawn_agent), and execute local scripts.
  • Sanitization: There is no evidence of sanitization or validation of the ingested specification content before it is interpolated into the agent's planning and execution logic.
  • [COMMAND_EXECUTION]: The documentation specifies the use of a local Node.js script for internal configuration management: node .claude/scripts/lib/workflow-skill-contract.cjs --write. This reflects a capability to execute localized scripts as part of the skill's maintenance cycle.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 01:34 AM