code-plan
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
jj diffcommand to inspect local code changes, which is a standard operation for its development-oriented purpose. - [EXTERNAL_DOWNLOADS]: It utilizes
WebSearchandWebFetchto retrieve external documentation and best practices, facilitating research-backed planning. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from processed data. 1. Ingestion points: Source code analyzed via
jj diffand external data fromWebSearchandWebFetch. 2. Boundary markers: No explicit delimiters or instructions are used to separate untrusted data from the agent prompt. 3. Capability inventory: Local command execution (jj diff), network access (WebFetch), and task delegation (spawning sub-agents). 4. Sanitization: No explicit sanitization or filtering of ingested content is performed.
Audit Metadata