plant-seed
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text for idea descriptions and rationales which are later used for codebase searches.
- Ingestion points: User input for idea action, rationale, and trigger conditions in SKILL.md.
- Boundary markers: Missing specific instructions to treat embedded content as untrusted data, though content is stored in a structured JSON file.
- Capability inventory: The skill uses Read, Write, Bash, Grep, Glob, and Edit tools as listed in SKILL.md frontmatter.
- Sanitization: No explicit sanitization or pattern-matching validation is mentioned for user input before it is written to the filesystem or passed to search tools.
- [COMMAND_EXECUTION]: Uses the Bash tool to execute mkdir -p .seeds/archived. This is a standard operation for directory management and is consistent with the skill's stated purpose of local data storage.
Audit Metadata