blocklet
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The instructions direct the agent to use the arc CLI for operations like arc blocklet recipe list, arc blocklet inspect, and arc blocklet build. These actions are necessary for the skill's documented purpose of managing blocklets.
- [INDIRECT_PROMPT_INJECTION]: The skill allows the agent to ingest and process data from local blocklet files, which could serve as a vector for indirect prompt injection.
- Ingestion points: arc blocklet list, arc blocklet inspect, and arc blocklet check commands read data from local paths (SKILL.md).
- Boundary markers: No explicit markers or instructions are provided to the agent to treat local blocklet data as untrusted or to ignore embedded instructions.
- Capability inventory: The skill allows shell command execution (arc CLI) and file system modifications (arc blocklet create, arc blocklet build) (SKILL.md).
- Sanitization: No sanitization or validation of the ingested file content is performed by the skill instructions.
Audit Metadata