swarm
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a local API server to determine the orchestration logic for its tasks.
- Ingestion points: The skill instructs the agent to fetch a 'ratified revision' of a plan from http://127.0.0.1:$PORT/api/plan and follow its sections (e.g., criteria, fanout) to guide its next actions.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the instructions for processing this external data.
- Capability inventory: The skill uses Bash and PowerShell to execute powerful binaries, performs local network requests via curl, and reads local configuration files.
- Sanitization: There is no evidence of sanitization for the plan sections or findings before they are incorporated into the agent's reasoning.
- [COMMAND_EXECUTION]: The skill makes extensive use of shell command execution to perform core functions such as server management, event pushing, and process control using the helix toolset.
Audit Metadata