demo-multi-agent-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation instructs the agent to execute shell commands to perform demonstrations, benchmarking, and testing.
- Evidence: Multiple examples in
SKILL.mdshow the usage ofpython tools/demo_multi_agent_workflow.pyto run scenarios and tests. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data within a multi-agent coordination pipeline, creating a vulnerability surface.
- Ingestion points: The skill is designed to ingest Pull Request content (via
--pr-number) and user-provided scenario configurations inSKILL.md. - Boundary markers: No delimiters or "ignore embedded instructions" warnings are documented for the data being processed.
- Capability inventory: The skill uses subprocess calls to execute a Python tool and coordinates multiple AI agents to perform complex tasks based on the input data.
- Sanitization: There is no documentation regarding the sanitization or validation of data retrieved from external sources like PRs before it is processed by the agents.
Audit Metadata