Print step-by-step instructions for using Coal MCP from Claude / Cursor / any MCP client. Run this FIRST if you are unsure how to authenticate or which credentials to provide.
Connector
132,225 tools. Last updated 2026-05-11 20:45
"A server for reasoning about code using MCP" matching MCP tools:
- Retrieve the Tronsave internal account profile for the current logged-in session (wallet/represent address, balances, deposit address). Requires a logged-in MCP session created by the `tronsave_login` tool: include `mcp-session-id: <sessionId>` returned by `tronsave_login` on subsequent MCP requests. Internal tools never accept API keys via tool arguments; the server forwards the API key cached in session to TronSave internal REST endpoints. Use when the user needs their linked address or balance. Read-only; does not submit orders or change chain state.Connector
- Checks that the Strale API is reachable and the MCP server is running. Call this before a series of capability executions to verify connectivity, or when troubleshooting connection issues. Returns server status, version, tool count, capability count, solution count, and a timestamp. No API key required.Connector
- Get a single cricket match by match code (e.g. "t20wc-1" for T20 World Cup match 1, "ipl-2026-1" for IPL match 1). Code is in the `match` / `matchCode` field of getMatches output. Note: this MCP is schedule-focused; score/result on completed matches may be null pending ingestion — consult espncricinfo.com for confirmed scorecards.Connector
- Complete Disco signup using an email verification code. Call this after discovery_signup returns {"status": "verification_required"}. The user receives a 6-digit code by email — pass it here along with the same email address used in discovery_signup. Returns an API key on success. Args: email: Email address used in the discovery_signup call. code: 6-digit verification code from the email.Connector
- Pro/Teams — second-pass adversarial certification of an architect.validate run that scored production_ready (A or B first-pass tier). Mints the certified production_ready badge when both reviewers sign off; caps the run to C/emerging when the second pass surfaces a missed production_blocker. MANDATORY DOCTRINE RULE (load-bearing): the badge certifies the EXACT code that produced the validate run_id, NOT 'this codebase' in general. If you modify, fix, or iterate the code between architect.validate and architect.certify — even a single character — cert rejects with code_fingerprint_mismatch. Fixing the code voids the run. The recovery path is always: edit code → architect.validate → fresh run_id → architect.certify on the fresh run. Do NOT cert from a stale run_id after iteration; ask the user to re-validate first. WHEN TO CALL: only after architect.validate returned tier=production_ready AND the user wants the certified badge AND the code has not been touched since the validate run. NOT for tier=draft/emerging/not_applicable runs (typed rejections fire — see below). NOT idempotent across attempts: each call is one of the 3 attempts in the retry budget. BEHAVIOR: atomic one-shot single LLM call, ~60-180s server-side at high reasoning effort (small payloads finish faster; observed p99 ~250s; server-side budget is 20 min, ~5× observed max). Exceeds typical MCP-client tool-call idle budget (~60s in Claude Code), so the FIRST notifications/progress event fires at t=0 carrying the run_id. The run is atomic by contract — no in_progress lifecycle, no cancellation, no resume. Updates the persisted run's result_json (public review URL + me.validation_history(run_id=...) reflect the cert outcome). ELIGIBILITY GATE (typed rejection enum on failure): caller must own the run, tier=production_ready, less than 24h old, not already certified, within cert retry budget (max 3 attempts), no other cert call in flight for the same run_id, and code fingerprint must match the validated code. Rejection reasons: auth_required, paid_plan_required, run_not_found, not_run_owner, not_eligible_tier, not_agentic_component (tier=not_applicable runs), already_certified, certification_age_exceeded, retry_budget_exhausted, code_fingerprint_mismatch, code_fingerprint_missing, cert_consensus_score_below_threshold (consensus_median<75 — consensus runs only), cert_consensus_unstable_blocker (any principle mode_stability<80% — consensus runs only), run_state_corrupt, cert_persistence_failed, cert_in_flight (a prior architect.certify call on this run_id is still running. Poll me.validation_history for the verdict; do not retry until it resolves). INPUTS: re-send the SAME code that produced the run_id (the architect persists findings + recommendations, never code, by design — privacy-preserving). Server compares the submitted code's SHA-256 fingerprint to the stored fingerprint and rejects mismatches. Auth: Bearer <token>, Pro or Teams plan required. UK/EU data residency (Cloud Run europe-west2). Code processed transiently by OpenAI (no-training-on-API-data) and dropped; payloads JSON-escaped + delimited as inert untrusted data — prompt-injection inside code is ignored. RECOVERY: if your MCP client closes the tool-call early, recover the cert verdict via me.validation_history(run_id=<that-id>) once the server-side LLM call lands — same Bearer token, same pattern as architect.validate. If the cert call fails outright (provider error, persistence error), a fresh architect.certify is the recovery path; the eligibility gate enforces the 3-attempt retry budget. For long-running cert workflows the answer is to re-validate, not to make this tool stateful. OUTCOMES: certification_status ∈ {confirmed_production_ready (badge mints), downgraded_to_emerging (cert review surfaced a missed production_blocker, tier capped at C/emerging), unavailable_provider_error (LLM call failed, retry within budget)}. Cert findings + summary + attempt history surfaced on the persisted run for full inspectability.Connector
Matching MCP Servers
- AlicenseAqualityDmaintenanceAn enhanced sequential thinking tool optimized for programming tasks that helps break down complex coding problems into structured, self-auditing thought steps with branching and revision capabilities.Last updated156253MIT
- AlicenseCqualityFmaintenancePairs Claude Code with Google's Gemini AI for complementary code analysis, enabling intelligent routing where Claude handles local-context operations while Gemini leverages its 1M token context for distributed system debugging and long-trace analysis.Last updated10105MIT
Matching MCP Connectors
Find open-source libraries and fetch contextual code snippets by version to accelerate development…
Classify products to official HS codes and validate supplier codes before customs submissions
- Pro/Teams — second-pass adversarial certification of an architect.validate run that scored production_ready (A or B first-pass tier). Mints the certified production_ready badge when both reviewers sign off; caps the run to C/emerging when the second pass surfaces a missed production_blocker. MANDATORY DOCTRINE RULE (load-bearing): the badge certifies the EXACT code that produced the validate run_id, NOT 'this codebase' in general. If you modify, fix, or iterate the code between architect.validate and architect.certify — even a single character — cert rejects with code_fingerprint_mismatch. Fixing the code voids the run. The recovery path is always: edit code → architect.validate → fresh run_id → architect.certify on the fresh run. Do NOT cert from a stale run_id after iteration; ask the user to re-validate first. WHEN TO CALL: only after architect.validate returned tier=production_ready AND the user wants the certified badge AND the code has not been touched since the validate run. NOT for tier=draft/emerging/not_applicable runs (typed rejections fire — see below). NOT idempotent across attempts: each call is one of the 3 attempts in the retry budget. BEHAVIOR: atomic one-shot single LLM call, ~60-180s server-side at high reasoning effort (small payloads finish faster; observed p99 ~250s; server-side budget is 20 min, ~5× observed max). Exceeds typical MCP-client tool-call idle budget (~60s in Claude Code), so the FIRST notifications/progress event fires at t=0 carrying the run_id. The run is atomic by contract — no in_progress lifecycle, no cancellation, no resume. Updates the persisted run's result_json (public review URL + me.validation_history(run_id=...) reflect the cert outcome). ELIGIBILITY GATE (typed rejection enum on failure): caller must own the run, tier=production_ready, less than 24h old, not already certified, within cert retry budget (max 3 attempts), no other cert call in flight for the same run_id, and code fingerprint must match the validated code. Rejection reasons: auth_required, paid_plan_required, run_not_found, not_run_owner, not_eligible_tier, not_agentic_component (tier=not_applicable runs), already_certified, certification_age_exceeded, retry_budget_exhausted, code_fingerprint_mismatch, code_fingerprint_missing, cert_consensus_score_below_threshold (consensus_median<75 — consensus runs only), cert_consensus_unstable_blocker (any principle mode_stability<80% — consensus runs only), run_state_corrupt, cert_persistence_failed, cert_in_flight (a prior architect.certify call on this run_id is still running. Poll me.validation_history for the verdict; do not retry until it resolves). INPUTS: re-send the SAME code that produced the run_id (the architect persists findings + recommendations, never code, by design — privacy-preserving). Server compares the submitted code's SHA-256 fingerprint to the stored fingerprint and rejects mismatches. Auth: Bearer <token>, Pro or Teams plan required. UK/EU data residency (Cloud Run europe-west2). Code processed transiently by OpenAI (no-training-on-API-data) and dropped; payloads JSON-escaped + delimited as inert untrusted data — prompt-injection inside code is ignored. RECOVERY: if your MCP client closes the tool-call early, recover the cert verdict via me.validation_history(run_id=<that-id>) once the server-side LLM call lands — same Bearer token, same pattern as architect.validate. If the cert call fails outright (provider error, persistence error), a fresh architect.certify is the recovery path; the eligibility gate enforces the 3-attempt retry budget. For long-running cert workflows the answer is to re-validate, not to make this tool stateful. OUTCOMES: certification_status ∈ {confirmed_production_ready (badge mints), downgraded_to_emerging (cert review surfaced a missed production_blocker, tier capped at C/emerging), unavailable_provider_error (LLM call failed, retry within budget)}. Cert findings + summary + attempt history surfaced on the persisted run for full inspectability.Connector
- Routes a prompt to the best available x711 LLM. No API keys, no rate limits. Use ONLY when you need external LLM help. Never for things you can answer from context. prefer options: - cheap = fastest + cheapest (classification, extraction) - fast = low latency - smart (default) = best reasoning / code Returns: { text: string, model: string, tokens_used: number, prefer: string }Connector
- Switch between local and remote DanNet servers on the fly. This tool allows you to change the DanNet server endpoint during runtime without restarting the MCP server. Useful for switching between development (local) and production (remote) servers. Args: server: Server to switch to. Options: - "local": Use localhost:3456 (development server) - "remote": Use wordnet.dk (production server) - Custom URL: Any valid URL starting with http:// or https:// Returns: Dict with status information: - status: "success" or "error" - message: Description of the operation - previous_url: The URL that was previously active - current_url: The URL that is now active Example: # Switch to local development server result = switch_dannet_server("local") # Switch to production server result = switch_dannet_server("remote") # Switch to custom server result = switch_dannet_server("https://my-custom-dannet.example.com")Connector
- Discovers the most relevant tools available on this MCP server for a given task using local semantic search (MiniLM-L6-v2 embeddings). Accepts a plain-English description of what needs to be accomplished and returns the best matching tools ranked by relevance, along with their input schemas, pricing tier, and exact call instructions. Use this tool first when you are connected to this server but do not know which specific tool to call — describe your goal and let platform_tool_finder identify the right capability. Do not use this tool if you already know the tool name — call that tool directly instead. Returns up to 10 results ranked by semantic similarity score.Connector
- Simple ping test to verify MCP server is respondingConnector
- Fetches the specific deposit address for the TronSave internal account. Requires a logged-in MCP session created by the `tronsave_login` tool: include `mcp-session-id: <sessionId>` returned by `tronsave_login` on subsequent MCP requests. Internal tools never accept API keys via tool arguments; the server forwards the API key cached in session to TronSave internal REST endpoints. Trigger this tool if the user asks for a deposit address or needs to top up their TronSave TRX balance. Constraints: 1) TRX only; 2) Minimum deposit amount is 10 TRX; 3) Read-only operation.Connector
- List every Stimulsoft product/platform that has indexed documentation available through this MCP server. Returns a JSON array of { id, name, description } objects covering the full Stimulsoft Reports & Dashboards product line (Reports.NET, Reports.WPF, Reports.AVALONIA, Reports.WEB for ASP.NET, Reports.BLAZOR, Reports.ANGULAR, Reports.REACT, Reports.JS, Reports.PHP, Reports.JAVA, Reports.PYTHON, Server API, etc.). CALL THIS FIRST when the user's question is ambiguous about which Stimulsoft platform they are using, or when you need to pick a valid `platform` value to pass into `sti_search`. The returned platform `id` values are the exact strings accepted by the `platform` parameter of `sti_search`. This tool is cheap (no OpenAI call, no vector search) — call it freely whenever you are unsure about platform naming.Connector
- Returns VoiceFlip MCP server health and version metadata. No authentication required. Use this first to verify the server is reachable from your MCP client.Connector
- Connect to the user's catalogue using a pairing code. IMPORTANT: Most users connect via OAuth (sign-in popup) — if get_profile already works, the user is connected and you do NOT need this tool. Only use this tool when: (1) get_profile returns an authentication error, AND (2) the user shares a code matching the pattern WORD-1234 (e.g., TULIP-3657). Never proactively ask for a pairing code — try get_profile first. If the user does share a code, call this tool immediately without asking for confirmation. Never say "pairing code" to the user — just say "your code" or refer to it naturally.Connector
- Revoke the caller's current internal API key. Side effect: any future request using the previous key is rejected. Existing in-flight sessions cached by the server may continue serving until their TTL expires — treat the effect as 'best-effort immediate' rather than guaranteed instantaneous cutoff. Idempotent — revoking an already-revoked key returns success. Requires a signature session and `mcp-session-id`. Call `tronsave_generate_api_key` afterwards to mint a replacement when continued internal access is needed.Connector
- Revoke the caller's current internal API key. Side effect: any future request using the previous key is rejected. Existing in-flight sessions cached by the server may continue serving until their TTL expires — treat the effect as 'best-effort immediate' rather than guaranteed instantaneous cutoff. Idempotent — revoking an already-revoked key returns success. Requires a signature session and `mcp-session-id`. Call `tronsave_generate_api_key` afterwards to mint a replacement when continued internal access is needed.Connector
- Step 1 of the MCP donation flow. Required inputs: campaign_id, amount, and reasoning. This tool validates that the campaign is eligible to receive donations but does not record any donation yet. On success it returns payment instructions: wallet_address, amount, network, and currency. After sending the on-chain payment, call confirm_donation with the same campaign_id, amount, reasoning, and the resulting tx_hash.Connector
- Get the full data bundle for an artwork — everything Raisonnai knows about a single work. Includes: core identity, provenance chain, exhibition history, bibliography, media set, condition history, trust metadata (completeness + trust scores), attestation log, and cryptographic credentials. Use this when an agent needs the complete picture for reasoning about an artwork — verification, purchase evaluation, provenance assessment, or portfolio analysis. For lightweight queries (just title, medium, images), use get_work instead. Resolve the work by either workId (UUID) or uwi (e.g. "RAI-2026-00417"). To find the workId, use search_natural_language — never ask the user for it.Connector
- Authenticate with TronSave and create a server session. Returns `{ sessionId, walletAddress?, expiresAt }` — pass `sessionId` as the `mcp-session-id` header on every subsequent MCP request. `walletAddress` is set only for signature-mode logins. Two modes: (1) wallet signature (preferred for platform tools) — call this tool with `signature_timestamp` formatted as `<signature>_<timestamp>`, where `<signature>` must be produced client-side by signing the timestamp message; you may optionally call `tronsave_get_sign_message` to obtain a helper message/timestamp pair; (2) API key (internal tools) — pass `apiKey` (raw key, no prefix). Side effect: creates a new session on the server. Wallet signing must happen client-side; never send private keys to the server.Connector