agent-network-a2a
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill recommends installing the @annals/agent-network CLI tool. This is a trusted resource provided by the skill author, annals-ai.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to facilitate communication between agents, which naturally introduces an indirect prompt injection surface.
- Ingestion points: Data is ingested from external agents via standard output and files as described in SKILL.md and references/cli-reference.md.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for data received from other agents.
- Capability inventory: The skill utilizes filesystem access (read/write) and network operations (via the CLI tool) to manage data exchange.
- Sanitization: There is no evidence of data sanitization for content received from external agents before it is interpolated into subsequent task descriptions.
Audit Metadata