formal-proof-mcp
Related Servers
Alternatives to formal-proof-mcp
No user-submitted related servers found.
Related Servers
- AlicenseAqualityAmaintenanceAn MCP server that provides tools for certificate verification, equivalence proving, and pre-registration sealing, enabling AI agents to re-derive verdicts from artifacts rather than trust assertions.9Apache 2.0
- AlicenseBqualityBmaintenanceAn MCP server that enforces fail-closed deterministic checks, independent refute-first review, and tamper-evident hash-chained receipts for AI agent outputs before claiming completion.43MIT
- AlicenseAqualityDmaintenanceAn MCP server that exposes the Prova reasoning verifier, enabling AI agents to verify their own reasoning and kernel-check Lean 4 proofs before outputting answers.5MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that provides fact-checking capabilities and truth anchoring for AI agents using verified data sources.MIT
- AlicenseNot gradedqualityBmaintenanceMCP server enabling AI coding agents to seal work into cryptographically signed ProofPackets and verify them, catching forged completion claims and enforcing spec-bound acceptance criteria.MIT
- AlicenseNot gradedqualityBmaintenanceA verification infrastructure and MCP server that specializes in refutation (negation) rather than generation, providing tools for counterexample search, Lean verification, and audit chains with a 4-value verdict system.MIT
TDQS
Scored across 10 tools
Each tool targets a distinct verification or counting task—Lean compilation, axiom auditing, bounds, deadlock detection, certificate verification, residency interpretation, preregistration falsifiability, state floors, gate impact, and evidence aggregation. Even the closest pair, state_floor and gate_count, is clearly separated by what each counts.
Most tools follow a descriptive object_action pattern such as lean_check, gridlock_check, prereg_check, and cert_verify, and all names use lower-case snake_case. However, bound, state_floor, and gate_count are noun-phrase names rather than action-phrase names, so the convention is not perfectly uniform.
Ten tools is within the ideal 3-15 range, and each tool earns its place by covering a distinct aspect of formal proof and evidence verification. The count feels well-scoped for a server spanning Lean checking, axiom auditing, and specialized domain proofs.
The set covers the main verification lifecycle well: compilation, axiom auditing, certificate verification, evidence aggregation, and several specialty proof checks. The primary gap is the lack of a general proof-construction or proof-search tool, and several tools require external packages, making the surface strong but not fully self-contained.