Tool Use.

Tool Use decides which Mickai-permissioned tool to invoke for a given step, drives the invocation, parses the response, and signs the entire round-trip into the audit ledger. Every tool call is gated by the Permissions brain and the operator's active clearance. Sensitive tool calls require a fresh voice-biometric match (patent 13). Per-tool rate limits cap the blast radius of any compromised credential (patent 03).

• 01Tool selection from the active scope
• 02Argument synthesis and validation
• 03Round-trip auditing with cryptographic signatures
• 04Voice-gated re-authentication on sensitive tools

Authoritative external corpora and standards this brain treats as canonical. Every retrieval against these sources is signed into the audit ledger so a regulator can prove which evidence drove which output.

• 01Mickai Patent 03, 12, 13, 20
• 02ReAct paper (Yao et al.)
• 03Toolformer and tool-use academic literature
• 04Model Context Protocol (MCP) specification
• 05OpenAPI 3.x specification
• 06gRPC and Protocol Buffers documentation
• 07ISO/IEC 25010 quality model
• 08JSON Schema specification
• 09Function-calling academic corpus

Mickai-native tooling primitives this brain implements internally. Codex for sovereign plain-text graph PKM, Lectern for spaced-repetition memory, Stele for citation-provenance, and domain-native primitives layered on top. No external services in the trust path; data stays on operator-personalised hardware.

• 01Sinew (tool-invocation framework)
• 02Aegis (per-tool policy enforcement)
• 03Wax-Seal (round-trip audit signing)
• 04Threshold (tool-credential gating)