inspiration-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the renku CLI and standard shell utilities to manage project state, list files, and persist analysis data. Evidence: Commands such as renku project open, renku inspiration show, and find are used as primary tools in SKILL.md and references/inspiration-analysis-cli-workflow.md.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata and file content, creating an attack surface for indirect prompt injection. \n
- Ingestion points: The agent reads folder names and image files from the local filesystem as part of the visual analysis process (SKILL.md). \n
- Boundary markers: No explicit boundary markers or safety instructions are provided to the agent to ignore potentially malicious instructions embedded in the analyzed images or folder names. \n
- Capability inventory: The skill executes shell commands via the Renku CLI and writes JSON documents back to the filesystem (SKILL.md, references/inspiration-analysis-cli-workflow.md). \n
- Sanitization: No sanitization or schema validation is performed on the input data before it is incorporated into the analysis findings.
Audit Metadata