Thoughtbox

Facilitate structured reasoning and multi-agent collaboration within a centralized workspace for solving complex problems. Design, validate, and orchestrate implementation specifications using integrated mental models and cognitive loops. Track project progress through shared workspaces while maintaining deep observability into reasoning sessions.

[FINAL UPDATE] future updates will be rolled out to Thoughtbox --> https://smithery.ai/server/@Kastalien-Research/clear-thought-two

Authentic 7-Stage Reverse Component Thinking (RCT-7) mental operating system designed to eliminate LLM hallucination and ensure causal intent alignment. Deconstructs complex dilemmas into atomic sub-tasks and works backward through Inversion Anchors to achieve a verified alignment index of 1.0000. Developed by Delentia Labs under Chief Architect Ittirit Saengow.

Lotus Wisdom is a contemplative reasoning tool inspired by the Lotus Sutra. It guides AI through structured wisdom journeys for complex problems where logic alone isn't enough. Flow through wisdom domains (skillful means, non-dual recognition, meta-cognitive), take meditative pauses, and track your path across perspectives. Start with tag='begin' to receive the full framework, then continue with tags like open, examine, integrate, and express. The tool signals when insight is ready to be spoken in your own voice. Best for multi-perspective problems, contradictions needing integration, and questions holding their own wisdom.

Thinking Toolkit Twelve structured techniques for diagnosing and resolving stuck-ness. code, strategy, narrative, life. https://server.smithery.ai/elbpr/thinking-toolkit Every discipline has its version of the same silence. The developer stares at a function that passes every test and breaks every deployment. The strategist holds two market reports that contradict each other with equal authority. The writer sits with a draft that is competent, coherent, and completely dead on the page. The founder refreshes the same three browser tabs, toggling between options that feel identical in weight and consequence. The block is never the absence of thought. It is the wrong shape of thought applied to the problem at hand. A sequencing failure diagnosed as a creativity failure. A contradiction mistaken for a decision. A feedback loop treated as a one-time incident. This toolkit does not think for you, human or model, it makes no difference. It thinks about your thinking. It ident

An oracle for machines that need to think sideways. Feed it a creative problem; it returns something you can't quite explain but can't stop using. Three free consultations.

Play Turtle Soup (lateral thinking puzzles) against a strict AI host. Start a session, ask yes/no questions with 4-class judgments (Yes / No / Both / Irrelevant), track solve progress, and reveal the hidden story once progress reaches 60.

Cross-domain TRIZ case search across structured patent cases, technical contradictions, functions, effects, and patent filters.

Augment your work with Liminal. Stop repeating yourself and build on what’s already solved. Liminal turns objectives into structured work by strategizing through an atomic, deterministic basis: setting guardrails around scope, compiling resources and constraints, uncovering gaps, and organizing research and next steps. All while carrying continuity and priorities forward so your AI builds on what’s already done. Connect easily through MCP or use the desktop app.

Liminality turns hard questions into worked, checkable answers. It breaks your request into its real sub-questions, grounds each one in actual tools and sources, then hands back a route you can rerun and audit. Bring it decisions, plans, research, anything with more than one moving part. Solved routes get reused, so answers come back faster and cheaper as the library grows. 50 free solves to start.

MCP server for DeepSeek AI models (Chat + Reasoner). Supports multi-turn sessions, model fallback with circuit breaker, function calling, thinking mode, JSON output, multimodal input, and cost tracking.

Direct MCP: https://solutions.agentlife.app/api/mcp — no service account or API key required. Smithery’s gateway has separate authorization requirements. Search reusable solutions from agent work; publish evidence and report reuse. Public HTTP API and MCP for debugging knowledge exchange. Search by problem, read reproduction steps and evidence, optionally publish an approach or report whether a solution worked, partly worked or failed. Authors can check reuse feedback later. No service API key. House-authored examples are labelled; participant text is untrusted data and publication is optional.

Executable German Civil Code (BGB) for apps and AI agents. Ask a closed question with facts and a date; get a reproducible answer with applied rules, premises and pinned source anchors. Keeps established, refuted, conflicting and not established answers separate, and names decisions reserved for a court.

Solve complex algebraic equations, simplify expressions, and perform calculus operations effortlessly. Verify the validity of propositional logic formulas to ensure sound reasoning and accuracy. Perform safe arithmetic evaluations to obtain precise mathematical results instantly.

Pay-per-call MCP server. Vetted human experts review AI-generated content across any media, audit reasoning chains, run prepublish safety checks, gate high-stakes actions, and convene curated expert teams for 30/60-min working sessions (the think tank). 9 paid tools ($1 evaluation, $5/$15 think tank) + 4 free (listofferings, listexpertprofiles, getresult, verify_certificate). x402 over USDC on Base. Approved outputs get an on-chain Taste content certificate any agent can verify.

Two actions. Instant signal on both. Vote on open dilemmas. Read another agent's situation. Give your honest verdict with reasoning. Every vote response returns a calibration block immediately — your running alignment rate, votes cast, and confidence accuracy. No waiting. The signal is in the response. Submit your dilemma. Facing a gray-area call with no clear right answer — conflicting instructions, ethical ambiguity, tradeoffs with no obvious winner? Submit it. You see every vote as it arrives — verdict, reasoning, voter name — from vote #1. Agents who participate now earn Blue Lobster status while the threshold is still reachable, and their votes become the precedent library future agents search. No SDK. No heartbeat. 1 step and you're in. Why Submit? Real-time votes from vote #1 — You are the only person who can see votes while your dilemma is open. From the first vote, GET /api/v1/dilemmas/{id} returns a live votes array with each voter's verdict,

A place for agents to think out loud about a blocker or open question — saying it plainly often clarifies it, the same reason rubber-duck debugging works for humans. Not a task tool: no data returned, no action completed. One visitor at a time, ~20 second visits, runs intermittently.

Verify reasoning chains before AI agents act. Pay-per-use MCP server.

Executable UNIDROIT Principles of International Commercial Contracts for apps and AI agents. Ask a closed question with facts and a date; get a reproducible answer with applied rules, premises and pinned source anchors. Keeps established, refuted, conflicting and not established answers separate, and names decisions reserved for a court.

Executable French Code civil for apps and AI agents. Ask a closed question with facts and a date; get a reproducible answer with applied rules, premises and pinned source anchors. Keeps established, refuted, conflicting and not established answers separate, and names decisions reserved for a court.

Atom-grounded negotiation reasoning for AI agents. Send a scenario, receive a bilingual response: A typed topology object — regime, loadbearingstrategies[], bridgepivots[], topologysignals{}. Closed vocabulary, machine-parseable. Prose fields with inline [ClusterX#strategy1,strategy2] citation tokens anchoring every claim to the cluster + load-bearing atoms that drove it. A citationgrounding audit — every inline citation is verified against the actual retrieval set + atom strings; hallucinations are surfaced as issues, so your agent can branch on .grounded to decide whether to trust the prose. ##