A CLI tool to discover, invoke, and cryptographically verify AI agents across multiple package registries and sources
Critical gaps across all dimensions. Tool definitions are inferred from main.go switch statement rather than explicitly registered with schemas. No input schemas visible in source code, all 8 tools lack documented parameter types and constraints. Descriptions are present but generic (10-80 chars, well below 50-200 char LLM-optimized baseline). No output schemas documented. No error handling guidance. No parameter validation rules. Tool names lack verb-noun clarity (e.g., 'invoke' is vague; 'invoke_agent' would be clearer). No evidence of idempotency, pagination, or composition patterns. Security concerns: FD_LIVE_KEY and GITHUB_TOKEN exposed as environment variables rather than server-side injection.
Runs a real search, resolves real entities, and shows a real, transparent confidence breakdown per resolved entity
Queries the real, local knowledge graph by name with case-insensitive substring matching
Invoke a real ForceDream agent
Real connector reputation for a known connector name
Search across real, available agent/server sources
Real, live health for every registered connector, plus real, locally-observed latency/reliability
Shows real, local telemetry state (level, install ID), plus a best-effort, real fetch of the backend's real, aggregate summary
No input schemas visible in source code. All 8 tools lack documented parameter types, constraints, and validation rules.
Tool definitions are inferred from main.go switch statement, not explicitly registered. No MCP tool registration visible.
No output schemas documented. LLMs cannot plan downstream calls or extract required fields. Violates pattern:tool and pattern:response-shaper.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 32 | 2026-07-28+ | v2 |
Verify a real Ed25519 proof, entirely client-side
Credentials (FD_LIVE_KEY, GITHUB_TOKEN) exposed as environment variables. Should use server-side secret injection. Violates pattern:secret-injection.
No error handling guidance. Tools lack recovery hints, error classification, or actionable error messages. Violates pattern:recovery-guide.