claat-creator
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch a linting tool from the author's GitHub repository (
github.com/nogikun/claat). - [REMOTE_CODE_EXECUTION]: The instruction to execute
go run github.com/nogikun/claat/cmd/claat-tools@v0.1.0 lintinvolves running a remote-sourced binary. This utility is provided by the skill author and is used for the primary function of document verification. - [INDIRECT_PROMPT_INJECTION]: The skill coordinates the processing of untrusted project data to generate manuals, which creates a potential surface for indirect prompt injection.
- Ingestion points: The skill description and workflow instructions specify ingesting "materials" such as project source code, README files, and notes from
SKILL.md. - Boundary markers: No explicit delimiters or boundary markers are instructed for use when the agent processes these external materials.
- Capability inventory: The skill environment facilitates file system operations and command execution (
go run) to process and validate the generated documentation. - Sanitization: There is no mention of input sanitization or filtering for the external project content before it is processed by the agent.
Audit Metadata