decide
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform file system operations to manage its data storage.
- In
SKILL.md(Step 6), the agent is directed to create a directory structure at${MAKERSKILLS_CONFIG:-$HOME/.config/makerskills}/decide/archive/if it does not already exist. - The skill defines a migration process requiring the agent to move files from a relative path (
references/decisions-archive/) to the newly created configuration directory. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface where user-provided answers are captured and stored.
- Ingestion points: User answers to specific triage questions are collected verbatim in Step 3 of
SKILL.md. - Boundary markers: The captured answers are delimited by Markdown headers (e.g.,
### <Question text>) within the archive template. - Capability inventory: The agent is granted capabilities for directory creation, file movement, and file writing to the local file system.
- Sanitization: There is no evidence of input validation or escaping for the user-provided answers before they are written to the decision log files.
Audit Metadata