plan-project

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input during the project discovery interview and incorporates it into a specification summary and implementation plan. These details are written to the filesystem in the docs/plans/ directory and linked within AGENTS.md. This represents a data flow where unvalidated external input is persisted into files that are intended to be read and potentially followed by AI agents in the future.
  • Ingestion points: User replies to discovery questions regarding product boundaries, data models, security models, and system integrations as defined in SKILL.md.
  • Boundary markers: The instructions do not define specific delimiters or warnings to be included in the generated files to distinguish user-provided content from agent instructions.
  • Capability inventory: The skill requires filesystem write access to create new directories, create Markdown files, and update the existing AGENTS.md file in the project root.
  • Sanitization: No explicit sanitization or validation logic is provided to filter user input before it is written to the documentation files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 02:25 PM
Security Audit — agent-trust-hub — plan-project