prd-to-plan
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes filesystem operations to create a
./plans/directory and write Markdown files. These operations are consistent with the tool's intended purpose of saving implementation plans locally. - [DATA_EXPOSURE]: The skill prompts the agent to explore the codebase to identify architectural patterns and data models. This involves reading local project files to provide context for the generated plan, which is a standard procedure for development-focused AI skills.
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external PRD content.
- Ingestion points: User-provided text or pointed files containing Product Requirement Documents (PRDs) in
SKILL.md. - Boundary markers: Absent; the instructions do not specify delimiters or warnings for the agent to ignore instructions embedded within the PRD.
- Capability inventory: The agent can read local files (codebase exploration) and write local files (plan generation in
./plans/). It does not have network access or administrative execution capabilities based on the provided instructions. - Sanitization: Absent; there is no mention of filtering or escaping content from the PRD before it is processed into the plan template.
Audit Metadata