technical-design

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill involves reading and processing untrusted data from the local workspace which could contain malicious instructions designed to influence the agent's design decisions.\n
  • Ingestion points: The agent reads .waypoint/local.yaml, feature-spec documents, repository instructions, and existing source code files (SKILL.md).\n
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in these files.\n
  • Capability inventory: The agent has the capability to read repository files and write Markdown design artifacts to the file system (SKILL.md).\n
  • Sanitization: No sanitization or validation of the ingested content is specified before it is incorporated into the design process.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing optional companion tools from a third-party GitHub repository.\n
  • Evidence: The documentation suggests using npx skills add mattpocock/skills --skill <name> to install additional capabilities from https://github.com/mattpocock/skills.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 01:06 AM
Security Audit — agent-trust-hub — technical-design