cs-feat-design

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands such as grep and project-specific Python utilities (validate-yaml.py, search-yaml.py) to search through project history and validate the structure of generated configuration files. These commands are integral to the skill's function and operate on local project data.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing external feature intent and roadmap files during the design drafting phase.
  • Ingestion points: The agent reads and processes data from {slug}-intent.md, {roadmap-slug}-roadmap.md, and {roadmap-slug}-items.yaml to populate design documents.
  • Boundary markers: No specific markers or delimiters are used to differentiate untrusted external content from the skill's internal instructions.
  • Capability inventory: The skill has permissions to write files to the local filesystem and execute internal project scripts for validation and search.
  • Sanitization: The skill employs validate-yaml.py to check the integrity of the resulting YAML checklists, though it does not explicitly sanitize the textual inputs from intent files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 05:37 AM