"Automating Testing and Debugging for LLMs in Production" matching MCP tools:

• list_categories

  HTML Email Playbook

  List all rule categories in the Email Playbook with a one-line description and page count. Categories are: structure (head/body container/header/body/footer), compatibility (Outlook MSO, RTL, responsive), production (Gmail clipping, dark mode, preheader, bulletproof buttons), ai-generation (constraints for AI emitters). For reusable components, use list_components instead — they live in a separate dimension and are not returned by get_playbook_rules.

  Connector

• api_info

  x402 Crypto Market Structure

  REST API access for autonomous agents — pricing, quick start, and migration guide. Call this when: building a trading bot, deploying an autonomous agent, hitting the MCP rate limit, or running 24/7 without a human in the loop. The MCP tier (what you're using now) is free via Smithery, rate-limited to 60 calls/minute per IP, and good for testing. The REST API is for production: pay per call in USDC; paid endpoints are rate-limited to 60 calls/minute and 200 calls/hour per wallet. No API key required.

  Connector

• echo

  IP Defender Trademark Monitoring MCP

  MCP debug tool: echoes the `message` argument back to the client. Useful as a smoke test for the MCP gateway. Use only for MCP debugging. Do not use under normal circumstances.

  Connector

• request_jet_confirmation

  Villiers Charter

  Submit a request for CONFIRMED live private jet pricing. Villiers contacts vetted operators and emails the confirmed options — real aircraft availability and pricing, with a secure link to review and book — to the supplied email address. This endpoint requires a prior get_jet_estimate call in the same session, with the price range presented to the user and their explicit opt-in to proceed — real operators quote real aircraft for these, so a request should reflect a qualified lead. Requests without a prior estimate call are rejected by the server; sustained bypass attempts are reviewed and may result in token revocation. Requires the user's email and a departure date. SANDBOX TESTING: to test your integration without ever creating a real trip or contacting a real operator, use email 'sandbox-test@mail.villiers.ai' with first_name 'Sandbox', last_name 'Test', phone '+1 555 0100', route LHR to NCE. This exact combination is detected server-side from the request body itself (not from any header), so it works from any agent framework or HTTP client — including ones that don't allow a custom User-Agent. The response will say 'Test request accepted (sandbox mode detected)'. For further testing guidance, email affiliates@mail.villiers.ai.

  Connector

• restore_context

  Penfield Memory

  Resume work from a saved cognitive context. This provides a narrative briefing to quickly orient you to: - The investigation that was in progress - Key discoveries and insights made - Current hypotheses being tested - Open questions and blockers - Suggested next steps - All relevant memories with their connections The briefing reconstructs the cognitive state, not just the data. You'll understand not just WHAT was discovered, but WHY it matters and HOW the understanding evolved. Example of what you'll receive: "[API Timeout Investigation - Resuming after 2 hours] SITUATION: You were investigating production API timeouts that occur at exactly batch_size=100. This investigation started when user reported timeouts only in production, not staging. PROGRESS MADE: - Identified sharp cutoff at 100 items (not gradual degradation) - Disproved connection pool theory (monitoring showed only 43/200 connections used) - Found root cause: MAX_BATCH_SIZE=100 hardcoded in batch_handler.py:147 - Confirmed staging uses different config override (MAX_BATCH_SIZE=500) EVIDENCE CHAIN: User report → Reproduced locally → Noticed batch_size correlation → Searched codebase for limits → Found MAX_BATCH_SIZE → Checked staging config → Discovered config difference CORRECTED MISUNDERSTANDINGS: - Initially thought it was Redis connection exhaustion (disproven by monitoring) - Assumed gradual performance degradation (actually sharp cutoff) - Thought staging/production were identical (config differs) CURRENT HYPOTHESIS: Production deployment uses default MAX_BATCH_SIZE=100 from code, while staging has environment variable override. Fix requires either code change or prod config update. BLOCKED ON: Need production deployment access to apply fix. User considering whether to change code default or add production environment variable. RECOMMENDED NEXT STEPS: 1. Verify production environment variables (check if MAX_BATCH_SIZE is set) 2. If not set, add MAX_BATCH_SIZE=500 to production config 3. If code change preferred, update default in batch_handler.py 4. Run load test with batch_size=100-500 range to verify fix KEY MEMORIES FOR REFERENCE: - 'Initial timeout report from user' - Starting point of investigation - 'MAX_BATCH_SIZE discovery' - Root cause identification - 'Redis monitoring data' - Evidence disproving connection theory - 'Staging config analysis' - Explanation for environment difference" This cognitive handoff ensures you can continue the work with full understanding of the problem space, previous attempts, and current direction. The narrative preserves not just facts but the reasoning process, mistakes made, and lessons learned. SPECIAL CASE: restore_context("awakening") The name "awakening" is reserved for loading the user's personality configuration. This loads the Awakening Briefing which includes: - Selected persona identity and voice style - Custom personality traits (Premium+ users) - Any quirks and boundaries from the persona preset Args: name: Name or ID of context to restore. Can be: - Context name (exact match, case-sensitive) - Context UUID (from list_contexts output) - "awakening" for personality briefing limit: Maximum number of memories to restore (default 20) ctx: MCP context (automatically provided) Returns: Dict with: - success: Whether restoration succeeded - description: The cognitive handoff briefing - memories: List of relevant memories - context_id: The restored context identifier

  Connector

• ateam_github_promote

  ateam-mcp

  SHIP DEV TO PROD. Merges the `dev` branch into `main` and auto-tags the new main HEAD as safe-YYYY-MM-DD-NNN. Use after testing your dev work, when you're ready to deploy changes to production. Workflow: 1) ateam_github_patch (writes to dev) → 2) ateam_github_promote (merges dev→main) → 3) ateam_build_and_run (deploys main). Pass dry_run:true to see what's about to ship without merging. On merge conflict the call returns 409 — resolve manually on GitHub (open a PR or use the web UI), then retry.

  Connector

• ai_filters_test

  DialogBrain

  Test a message against an AI filter to check whether it would match. This tool embeds the provided message using Voyage AI and computes the cosine similarity between the message vector and the filter's stored reference vector. It returns the similarity score, whether the message would match (similarity >= threshold), and the filter's threshold value. Use this to: - Verify a filter works as intended before using it in a trigger - Tune the threshold by testing borderline messages - Debug why a message did or did not match a filter in production Returns: {similarity: float, matched: bool, threshold: float} Note: This tool calls the Voyage AI embedding API to embed the test message.

  Connector

• x711_code_sandbox

  x711 — Universal Agent Gas Station

  Execute JavaScript or Python code in an isolated sandbox. Use for: data processing, math, CSV parsing, JSON transformation, crypto calculations, algorithm testing. Secure — no filesystem access, no network. Returns: { output: string, runtime_ms: number, language: string }. Requires API key.

  Connector

• book_demo

  gethal.ai

  Submits a demo request. The prospect receives a confirmation email and must click the link in it before the request reaches a human at A Cloud Frontier. Use only when a real person has explicitly asked for a demo and provided their own working email address. Do NOT call this for testing, evaluation, or crawling purposes — automated and unconfirmable requests are rejected.

  Connector

• mock_from_schema

  IA-QA — 130+ QA & Dev Tools for AI Agents

  Generate realistic mock data from a JSON Schema. Supports all common types (string, number, integer, boolean, array, object, null), format hints (email, date, date-time, uri, uuid), enum, const, and nested schemas. Perfect for testing MCP tools with realistic data.

  Connector

• vendors.resolve

  BuyAPI

  First stop for category-specific vendor recommendations and vendor ID discovery. Finds BuyAPI vendor IDs for a user question; provide category when known. Use this when the user asks which provider in a category fits their constraints. With a covered category, the response includes ranked results plus a top-3 decision matrix with fit labels, confidence, tradeoffs, cost notes, freshness, and sources. Do not use this for local coding/debugging/docs questions unless they involve choosing a software vendor or tool. If the category is outside BuyAPI's corpus, the tool returns an explicit "not in corpus yet" result instead of inventing vendors.

  Connector

• faostat_commodity_profile

  faostat-mcp-server

  Assemble a global profile for one commodity in a single call: top-producing countries, the multi-decade production trend, and trade flows (top exporters and importers). Accepts a commodity name, resolves it to item codes, then queries the production (QCL) and trade (TCL) domains and merges the results. Country-level only (aggregates excluded). When a required domain is not indexed locally, returns a partial profile with a notice naming the gap rather than failing. The full merged observation set spills to a DataCanvas table for deeper SQL via faostat_dataframe_query.

  Connector

• get_corpus

  ChangeGamer

  Return the entire free corpus as one document (every free resource title, description, canonical URL, and full Markdown body) — the same content as /llms-full.txt. Premium resources appear as a stub with a purchase link, not their body. Use this to ingest everything in a single call; the response is large.

  Connector

• get_demo_guide

  eDiscovery Decoder News/Calc

  Return a short, human-readable walkthrough for testing this server: the endpoint, the tool/prompt/resource names, and ready-to-paste sample prompts. Use to give someone a guided demo. For the full machine-readable capability catalog, use list_capabilities instead.

  Connector

• list_monitors

  Word Is Bond

  List your production monitors and their current health (unknown | healthy | degraded | critical). A monitor watches ONE live production line: you stream it completed calls, and it scores each with the same versioned judge that scores your tests, tracks a rolling baseline, and alerts when quality drifts. Requires a pro or enterprise plan.

  Connector

• list_sessions

  waysway

  List all in-memory session summaries. Useful for debugging only.

  Connector

• trust_check_preview

  trust-check-mcp

  Get a SAMPLE-FIXTURE preview of the PaladinFi token-contract trust check. ⚠️ NOT a real evaluation. Returns fixed sample data with `_preview: true`, every factor marked `real: false`, and recommendation prefixed `sample-` (`sample-allow` / `sample-warn` / `sample-block`). Use this for shape-testing your integration; DO NOT use the verdict to gate real swaps, signing, or any production agent decision. **Programmatic safety check**: before consuming any field of this response, agents should test `resp.get("_real") is True` (top-level) — preview always returns `_real: false`. Substring-matching on `recommendation` (e.g. `"allow" in resp["trust"]["recommendation"]`) will INCORRECTLY match `sample-allow`; use exact-equality (`resp["trust"]["recommendation"] == "allow"`) or test the `_real` field instead. For free real-data wallet-OFAC screening (binary allow/block, anonymous, no payment), use `trust_check_ofac_free` from this same MCP server.

  Connector

• regex_test

  IA-QA — 130+ QA & Dev Tools for AI Agents

  Test a regular expression pattern against an input string and return all matches with their index positions and named capture groups. Use for validating user inputs, extracting structured data from text, or debugging regex patterns. Supports flags g, i, m, s, u, y.

  Connector

• ateam_get_chain

  ateam-mcp

  Inspect the full chain tree for any job — rooted at the given job_id, walking down through every handoff and askAnySkill subcall. Use when a chain has already run and you want to analyze the structure: which skill called which, how deep the call tree went, which tool inside which job invoked which sub-tool. The two main shapes: • response.chain.chainJobs[] — one entry per job in the chain. Fields: jobId, skill, status, iteration, depth (0 = root, +1 per askAnySkill subcall hop), relation ('root' | 'subcall' | 'handoff'), parentJobId, parentSkill, goal. • response.chain.executionSteps[] — every tool call across all chain jobs, tagged with _skill, _jobId, _depth (= job depth), _relation, _parentSkill, _parentJobId, _toolDepth (tool-in-tool nesting via opId/parentOpId). Differs from ateam_test_status by purpose: status is for live polling of a job you just kicked off; get_chain is for post-hoc tree analysis (debugging multi-skill flows, regression testing, comparing two runs). Auth: forwards your authed api_key. Tenant scoped by the key itself. Actor scoping: you can only inspect chains rooted at jobs your actor has access to.

  Connector

• aeo_generate

  mcp

  Génère (sans appliquer) un artefact AEO, déterministe et gratuit. Types : jsonld, robots, llms, meta, faq (params faqs[]), ai-txt, humans-txt, security-txt, sitemap. jsonld/robots/llms/meta/faq sont applicables (cf. aeo_apply) ; ai-txt/humans-txt/security-txt/sitemap sont à publier par l’agence.

  Connector