planify
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from the local workspace and the web, which could contain instructions intended to influence the agent's behavior.
- Ingestion points: Processes user arguments, project documentation (CLAUDE.md, AGENTS.md), existing source code, and external content retrieved via WebFetch.
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in the analyzed data, though it does include a mandatory human-in-the-loop validation step in Phase 3.
- Capability inventory: The agent can read files, perform network searches, and invoke secondary skills to write files to the disk.
- Sanitization: No explicit sanitization or content filtering is implemented for ingested data.
- [EXTERNAL_DOWNLOADS]: The skill utilizes WebSearch and WebFetch to look up design patterns and best practices. These network operations are aligned with the skill's primary purpose of technical planning and do not involve sensitive data exfiltration.
- [COMMAND_EXECUTION]: The skill performs directory traversal and file searches using Glob and Grep tools to map project structure and conventions. These operations are limited to the current workspace.
Audit Metadata