ck
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local
gitcommands (log, diff, rev-list) usingspawnSyncto monitor project activity and summarize session changes. These commands are executed with structured arguments and are limited to the project directory.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it parses content from project-controlled files likeCLAUDE.md,package.json, andREADME.md. If these files contain malicious instructions, they could be parsed and displayed to the agent during initialization or session start. - Ingestion points:
commands/init.mjsandhooks/session-start.mjsread and parse documentation and metadata files from the project root. - Boundary markers: No specific delimiters or "ignore instructions" markers are applied to the ingested content before it is presented to the agent.
- Capability inventory: The skill can perform file system writes to its internal storage directory and execute local
gitcommands. - Sanitization: The skill does not perform sanitization or validation of the text content extracted from project files before processing it.
Audit Metadata