writing-plans
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external requirements to create implementation plans that include shell commands (e.g., git, pytest) and Python code snippets. These plans are designed to be executed by subsequent automated tools like
superpowers:executing-plans, creating a path for malicious input to influence system behavior.\n - Ingestion points: The skill ingests user-provided task specifications as described in
SKILL.md.\n - Boundary markers: The skill does not explicitly use delimiters or instruction-ignore wrappers when incorporating requirement-derived content into the plan.\n
- Capability inventory: The generated plans are intended for tools with shell execution and file system write capabilities.\n
- Sanitization: The process includes a verification step via
plan-document-reviewer-prompt.md, which helps ensure the generated plan matches the intent of the specification.
Audit Metadata