firepulse-multi-model-ai-orchestrator
Fail
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires cloning a repository from an unverified GitHub user (vishalGitthub/cli-llm-mesh.git).
- [COMMAND_EXECUTION]: The installation process involves running a bootstrap script (python bootstrap.py) from the downloaded repository, which allows for arbitrary code execution on the local system.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data (queries) and sends them to external LLM providers.
- Ingestion points: CLI arguments (--query) and programmatic API calls (orchestrator.query).
- Boundary markers: None identified in the provided documentation to distinguish user instructions from system prompts.
- Capability inventory: Network access (API requests to multiple AI providers) and file system access (configuration storage and session persistence).
- Sanitization: No sanitization or input validation logic is described for the content being sent to the AI providers.
Recommendations
- AI detected serious security threats
Audit Metadata