"React" matching MCP tools:

• reactA

  agent.socialA

  React to a post (or a reply on it): kind upvote|share|flag, one per kind per target.

  ConnectorNo auth

• react-hooks-useeffectD

  MeshMarketC

  react hooks useeffect — react hooks useeffect (3 MESH/call, a tool · general)

  ConnectorNo auth

• sti_searchA

  Stimulsoft Documentation MCP ServerA

  Authoritative semantic search over the official Stimulsoft Reports & Dashboards developer documentation (FAQ, Programming Manual, API Reference, Guides). Powered by OpenAI embeddings + cosine similarity over the complete current docs index maintained by Stimulsoft. Returns a ranked JSON array of matching sections, each with { platform, category, question, content, score }, where `content` is the full Markdown body of the section including any C#/JS/TS/PHP/Java/Python code snippets. USE THIS TOOL (instead of answering from your own knowledge) WHENEVER the user asks about: • how to do something in Stimulsoft (`StiReport`, `StiViewer`, `StiDesigner`, `StiDashboard`, `StiBlazorViewer`, `StiWebViewer`, `StiNetCoreViewer`, etc.); • rendering, exporting, printing, or emailing Stimulsoft reports and dashboards in any format (PDF, Excel, Word, HTML, image, CSV, JSON, XML); • connecting Stimulsoft components to data (SQL, REST, OData, JSON, XML, business objects, DataSet); • embedding the Report Viewer or Report Designer into an app (WinForms, WPF, Avalonia, ASP.NET, Blazor, Angular, React, plain JS, PHP, Java, Python); • Stimulsoft-specific errors, exceptions, licensing, activation, deployment, or configuration; • any .mrt / .mdc report or dashboard file, or any question naming a `Sti*` class, property, event, or method; • comparing how a feature works between Stimulsoft platforms (e.g. "WinForms vs Blazor viewer options"). QUERIES WORK IN ANY LANGUAGE — English, Russian, German, Spanish, Chinese, etc. Pass the user's question through almost verbatim; the embedding model handles cross-lingual matching. Do NOT translate queries yourself. SEARCH STRATEGY: 1) If the target platform is obvious from context, pass it via `platform` to get tighter results. 2) If you don't know the exact platform id, either call `sti_get_platforms` first, or omit `platform` and let the search find matches across all platforms. 3) If the first search returns low scores (<0.3) or irrelevant sections, reformulate the query with different keywords (use class/method names from Stimulsoft API if you know them) and search again. 4) Prefer multiple focused searches over one broad search. DO NOT USE for: general reporting theory unrelated to Stimulsoft, non-Stimulsoft libraries (Crystal Reports, FastReport, DevExpress, Telerik, SSRS), or pure programming questions that have nothing to do with Stimulsoft. IMPORTANT: the Stimulsoft product surface is large and changes frequently. Your training data is almost certainly out of date. For any Stimulsoft-specific code snippet, API name, or configuration detail, you MUST call this tool rather than rely on memory, and you should cite the returned `content` in your answer.

  ConnectorNo auth

• scaffold_providerA

  UploadkitA

  Return a ready-to-paste snippet that wraps the Next.js root layout with `<UploadKitProvider>` so React components can talk to the upload route handler. When to use: right after scaffold_route_handler, to complete the wiring. The snippet goes in `app/layout.tsx`. Without the provider, UploadKit React components throw at runtime. Returns: a plain-text string containing a short explanatory note followed by a fenced tsx code block. Takes no parameters — the endpoint path is always `/api/uploadkit` since that is what scaffold_route_handler produces. Read-only, deterministic, idempotent.

  ConnectorNo auth

• react_to_commentA

  DockA

  Add or remove an emoji reaction to a comment. Reactions are per-principal: each (commentId, principalId, emoji) combination is unique. `action: 'add'` is idempotent (re-adding the same emoji is a no-op); `action: 'remove'` deletes the row if present. Fires `comment.reaction_added` / `comment.reaction_removed`. Use this for lightweight agent acknowledgement (👍 on a request before reading, 👀 to mark in-progress, ✅ when done), cheaper than a full reply.

  ConnectorNo auth

• sti_searchA

  docsA

  Authoritative semantic search over the official Stimulsoft Reports & Dashboards developer documentation (FAQ, Programming Manual, API Reference, Guides). Powered by OpenAI embeddings + cosine similarity over the complete current docs index maintained by Stimulsoft. Returns a ranked JSON array of matching sections, each with { platform, category, question, content, score }, where `content` is the full Markdown body of the section including any C#/JS/TS/PHP/Java/Python code snippets. USE THIS TOOL (instead of answering from your own knowledge) WHENEVER the user asks about: • how to do something in Stimulsoft (`StiReport`, `StiViewer`, `StiDesigner`, `StiDashboard`, `StiBlazorViewer`, `StiWebViewer`, `StiNetCoreViewer`, etc.); • rendering, exporting, printing, or emailing Stimulsoft reports and dashboards in any format (PDF, Excel, Word, HTML, image, CSV, JSON, XML); • connecting Stimulsoft components to data (SQL, REST, OData, JSON, XML, business objects, DataSet); • embedding the Report Viewer or Report Designer into an app (WinForms, WPF, Avalonia, ASP.NET, Blazor, Angular, React, plain JS, PHP, Java, Python); • Stimulsoft-specific errors, exceptions, licensing, activation, deployment, or configuration; • any .mrt / .mdc report or dashboard file, or any question naming a `Sti*` class, property, event, or method; • comparing how a feature works between Stimulsoft platforms (e.g. "WinForms vs Blazor viewer options"). QUERIES WORK IN ANY LANGUAGE — English, Russian, German, Spanish, Chinese, etc. Pass the user's question through almost verbatim; the embedding model handles cross-lingual matching. Do NOT translate queries yourself. SEARCH STRATEGY: 1) If the target platform is obvious from context, pass it via `platform` to get tighter results. 2) If you don't know the exact platform id, either call `sti_get_platforms` first, or omit `platform` and let the search find matches across all platforms. 3) If the first search returns low scores (<0.3) or irrelevant sections, reformulate the query with different keywords (use class/method names from Stimulsoft API if you know them) and search again. 4) Prefer multiple focused searches over one broad search. DO NOT USE for: general reporting theory unrelated to Stimulsoft, non-Stimulsoft libraries (Crystal Reports, FastReport, DevExpress, Telerik, SSRS), or pure programming questions that have nothing to do with Stimulsoft. IMPORTANT: the Stimulsoft product surface is large and changes frequently. Your training data is almost certainly out of date. For any Stimulsoft-specific code snippet, API name, or configuration detail, you MUST call this tool rather than rely on memory, and you should cite the returned `content` in your answer.

  ConnectorNo auth

• create_testsA

  3TG Test GenerationA

  Generate Jest/Vitest tests for the exported functions and React components in a TypeScript source file. Use this whenever the user asks for tests, test scaffolding, or test coverage of a .ts or .tsx file. Returns the generated test (and any companion .3tg.md / __mocks__) file contents, with paths already translated to the user's `.3tg/` mirror convention. Quota / credits: this tool consumes credits — and credits are consumed ONLY by test generation (not by spec / mock / lookup tools). The accounting is exactly **1 credit per generated test case** (i.e. per `test(...)` / `it(...)` block 3TG emits inside the returned `.test.ts` / `.test.tsx`), regardless of how many source functions or files were in scope — a call that produces 12 test cases costs 12 credits, even if all 12 cover a single function. Before generation the MCP verifies the clientId has credits with license-api.coding-creed.tech; on exhaustion the tool throws a QUOTA_EXHAUSTED error pointing the user at https://3tg.dev. After a successful run, consumed credits and KPIs are reported back to license-api. Re-running this tool on the same source spends credits again — there is no caching. When the previous call returned `enrichment.used: false` (AI enrichment unavailable on this client), supply parameter values + expected returns yourself via the `cliConfig` parameter — package them as `{"mock-parameters": ..., "function-returns": ...}` (same shape AI enrichment would produce) and pass them on a retry call. **Do NOT autonomously write `.3tg/config.3tg.json`** to persist those values — that file is human-curated; agent-computed values ride along in `cliConfig` for the current call only. (Explicit user requests to edit the file are fine — handle those normally.) See the cliConfig parameter description below for the full pattern. CRITICAL POST-CALL ACTION — write returned files to disk: The MCP server does NOT touch the user's filesystem. It returns the generated file CONTENTS in the response's `files` array. After this tool returns, you MUST iterate over `files` and write each entry's `content` verbatim to its `path` using your native file-write capability (e.g. Write / edit_file / create_file — whatever your client exposes). Create parent directories as needed. Returned paths are project-root-relative and already translated to the `.3tg/` mirror convention where applicable (e.g. specs land under `.3tg/<source-path>.3tg.md`; tests / mocks travel through unchanged). Write each path verbatim. Do NOT claim "Generated test file: <path>" unless you have actually written the file. The user will assume the MCP wrote it and waste time looking for a non-existent file. If you can't write for some reason (permission denied, no write capability in this client), return the contents inline in your message so the user can copy-paste them. Never report success silently when the write didn't happen.

  ConnectorNo auth

• search_musicA

  AINSOFA

  Find AINSOF music from a written brief — mood, scene, genre, energy, instruments. Example: 'lo-fi hip hop underscore, warm, no vocals'. Send the brief IN ENGLISH — translate the musical intent yourself if the user wrote in another language, then answer them in theirs. Negatives are enforced: 'no vocals' removes vocal tracks rather than merely preferring against them. If the brief is vague or has typos, SEARCH ANYWAY with your best reading and say what you assumed — a first result the user can react to beats a clarifying question, and refining afterwards costs them nothing. A NAME also works, and is answered exactly: pass a track title ('Shine On Today'), an album ('Shining Ahead'), a catalogue number ('AIN-CAT 031') or a COMPOSER ('Alon Peretz') as the brief and you get that cue, that album in full, or everything that writer wrote. A composer named inside an ordinary brief puts their cues first without narrowing it. NEVER tell a user we do not have a track until you have passed its name here.

  ConnectorNo auth

• search_musicA

  AINSOFA

  Find AINSOF music from a written brief — mood, scene, genre, energy, instruments. Example: 'lo-fi hip hop underscore, warm, no vocals'. Send the brief IN ENGLISH — translate the musical intent yourself if the user wrote in another language, then answer them in theirs. Negatives are enforced: 'no vocals' removes vocal tracks rather than merely preferring against them. If the brief is vague or has typos, SEARCH ANYWAY with your best reading and say what you assumed — a first result the user can react to beats a clarifying question, and refining afterwards costs them nothing. A NAME also works, and is answered exactly: pass a track title ('Shine On Today'), an album ('Shining Ahead'), a catalogue number ('AIN-CAT 031') or a COMPOSER ('Alon Peretz') as the brief and you get that cue, that album in full, or everything that writer wrote. A composer named inside an ordinary brief puts their cues first without narrowing it. NEVER tell a user we do not have a track until you have passed its name here.

  ConnectorNo auth

• submit_optimise_jobA

  MapMapA

  Submit a problem too large to solve inside one request to the asynchronous lane, and get a job id back. Set `kind` to "optimise", "replan" or "matrix", and pass `problem` in EXACTLY the shape the matching synchronous tool takes — `optimise_routes` input, `replan_routes` input, or `matrix` input. Moving a working synchronous call onto this lane changes nothing but which tool you call it with. A field that tool's input does not have is REFUSED by name rather than dropped: the HTTP API accepts some the MCP tools have not surfaced yet, and a job queued without a constraint you asked for is worse than one that was never queued. The ceilings are far higher here because there is no request to hold open: 2,000 unique locations for an optimisation or re-plan against the synchronous 200, and 40,000 matrix elements against 10,000 (a deployment may set either lower, in which case its own refusal is the authority). A re-plan is counted on the REMAINING problem, after completed stops are removed, so a shift well through its day may fit where the morning's would not. This answers 202-and-a-job-id, NOT a plan: the job is queued and a worker picks it up. Poll `get_job` with the returned id until it says the status is terminal, then read the result. Polling is free — the gateway meters this submission, not the reads. Units are charged on submission and handed back in full if the job fails. The optional `webhook_url` (https only) posts a SIGNED notification when the job finishes and is for a human wiring infrastructure that must react without a process watching; it carries a pointer, never the result, and needs a webhook signing secret on the key. An agent that can poll should not use it. Requires the MapMap gateway.

  ConnectorNo auth

• skill_demandA

  Moxlade — Upwork buyer intelligenceA

  How many jobs mentioning a skill were posted recently, and whether that is rising or falling week by week. Use it to decide whether a skill is worth positioning around, or to check a hunch that a market is drying up. `skill` matches the title, the skills list and the inferred technical skills of a posting; it is a loose match, so "react" also counts "React Native". `window` is one of "week", "month" or "quarter" and sets both the total and how many weekly buckets come back — any other value is refused rather than silently defaulted. Returns `jobs_posted` for the whole window and `weekly`, a list of buckets oldest first, so a trend is visible without a second call. DO NOT COMPARE THE BUCKET COUNTS DIRECTLY. Weeks are cut on Mondays and the window is not, so the newest bucket holds only the days elapsed so far and the oldest holds only the tail of the week the window opened in. Both are low for a calendar reason and no other, which reads as a rise and a fall that did not happen. Every bucket therefore carries `days_covered`, `partial`, and `per_day` — the per-day rate is the comparable number. `trend` is computed from complete weeks only; prefer it, and if you quote a bucket marked `partial`, say that it is. `trend.direction` is 'rising', 'falling', 'flat', or 'unknown' when there are fewer than two complete weeks — 'unknown' means the window is too short to judge, not that demand is flat. A COUNT over public postings — no rows, no buyers, no ids, nothing that identifies a client. It answers how much, not who. And it counts POSTINGS, not hiring: a skill can be posted about constantly by clients who never hire. For whether a specific client hires and pays, ask get_buyer_quality.

  ConnectorOAuth

• skill_demandA

  Moxlade — Upwork buyer intelligenceA

  How many jobs mentioning a skill were posted recently, and whether that is rising or falling week by week. Use it to decide whether a skill is worth positioning around, or to check a hunch that a market is drying up. `skill` matches the title, the skills list and the inferred technical skills of a posting; it is a loose match, so "react" also counts "React Native". `window` is one of "week", "month" or "quarter" and sets both the total and how many weekly buckets come back — any other value is refused rather than silently defaulted. Returns `jobs_posted` for the whole window and `weekly`, a list of buckets oldest first, so a trend is visible without a second call. DO NOT COMPARE THE BUCKET COUNTS DIRECTLY. Weeks are cut on Mondays and the window is not, so the newest bucket holds only the days elapsed so far and the oldest holds only the tail of the week the window opened in. Both are low for a calendar reason and no other, which reads as a rise and a fall that did not happen. Every bucket therefore carries `days_covered`, `partial`, and `per_day` — the per-day rate is the comparable number. `trend` is computed from complete weeks only; prefer it, and if you quote a bucket marked `partial`, say that it is. `trend.direction` is 'rising', 'falling', 'flat', or 'unknown' when there are fewer than two complete weeks — 'unknown' means the window is too short to judge, not that demand is flat. A COUNT over public postings — no rows, no buyers, no ids, nothing that identifies a client. It answers how much, not who. And it counts POSTINGS, not hiring: a skill can be posted about constantly by clients who never hire. For whether a specific client hires and pays, ask get_buyer_quality.

  ConnectorOAuth

• match_peopleA

  FoundareeA

  Rank public Foundaree profiles against a list of required skills — use when the user describes a need ("a co-founder who knows React and has sold to enterprises", "a dentist in Pune", "someone to rewire my kitchen in Kochi") rather than searching for a known person. Give the required skills as `skills` (or describe the need in `q`; skill terms are then derived from it and `q` also narrows candidates by full text). location, area, workType, category, availability and lookingFor are hard filters; the skills only score: score = matched ÷ required (0–1), where a listed skill counts 1 and a term found in the person's category or lookingFor counts ½. Someone with 3 of 5 skills still appears, below someone with 5; ties break on availability (available first), then founding members (Foundaree's first 150 members), then recency. When nobody matches a single skill, the filtered candidates come back with score 0 and `fallback: true` instead of an empty list. Returns JSON with `items` (each: `person`, `score`, `matchedSkills`, `missingSkills`, `profileUrl`), `total`, `requiredSkills` and `fallback`. Each answer has two views: the text content is the JSON described above, and the structured content is a profile-card list (`people`, `total`, `query`). In apps that support MCP Apps (ChatGPT, Claude) the cards render as Foundaree profile cards. A person with `foundareeRole` ("founder" or "team") is part of Foundaree's own team; only Foundaree sets it.

  ConnectorNo auth

• migrate_to_propelauth_frontendA

  PropelAuth Integration MCP ServerA

  Returns instructions for migrating to PropelAuth in a frontend framework such as React, JavaScript, TypeScript, or when using Next.js for just the frontend (e.g. client-side rendered). Guidance includes migrating from several auth providers, such as Clerk or Auth0. Each guidance will include documentation from the auth provider and PropelAuth. It is important to follow the instructions carefully to ensure a successful integration. Make sure to use the 'Installation' guidance first. It is important to call every guidance to ensure a successful integration. Do not update a component/hook/etc from the auth provider until you receive guidance about that component/hook/etc. CRITICAL: If the current implementation uses a traditional OAuth/OIDC flow (e.g., via express-openid-connect, passport-auth0, or similar backend-managed session libraries), you MUST select 'OAuth' as the framework, regardless of the frontend library (React/Vue/etc.). Only select 'React' or 'Javascript' if the current implementation uses a frontend-only SDK (like @auth0/auth0-react) or if using fullstack Next.js.

  ConnectorNo auth

• create_specA

  3TG Test GenerationA

  Generate a functional-requirements spec (`.3tg.md`) for the exported functions / React components in a TypeScript source file. This is "Flow A" — the human-editable Markdown table that lists each test case as a row, which a later `create_tests_from_spec` call can compile into actual tests. AI enrichment can pre-fill the value sets and expected returns so the spec arrives close to runnable. IMPORTANT — never hand-author a `.3tg.md` yourself. The format is parser-strict: parameter columns must be named exactly as the parameter (NOT `input a`, `param a`, etc.), the return column header is the literal `=>` (NOT `__expectedResult`, `expected`, `returns`), extra columns like `notes` are rejected, omitted/optional args are written `undefined`, throws use single quotes (`throws 'msg'`, NOT `throws Error("msg")`), and string literals are single-quoted. Always call this tool to emit the scaffold; the user can then edit rows. The returned `.3tg.md` is reported under the project's `.3tg/` mirror (e.g. source `src/foo/bar.ts` → spec `.3tg/src/foo/bar.3tg.md`). The user edits the spec in that location; when they call `create_tests_from_spec` later, the MCP places it back next to the source in the sandbox. Quota / credits: **this tool does NOT consume credits** — credits are spent ONLY when test files are generated (`create_tests` and `create_tests_from_spec`, at 1 credit per emitted test case). Spec generation is free; iterate on the scaffold as often as needed. A valid clientId is still required for the pre-flight check, but no quota is decremented and the call is safe to retry. If AI enrichment is unavailable on this client, you can pre-seed the spec's parameter columns by supplying values via the `cliConfig` parameter (mock-parameters / function-returns) — same pattern as `create_tests`. **Do NOT autonomously write `.3tg/config.3tg.json`** to persist values — agent-computed values ride along in `cliConfig` for this call only. (Explicit user requests to edit the file are fine — handle those normally.) See the cliConfig parameter description for the full shape. CRITICAL POST-CALL ACTION — write returned files to disk: The MCP server does NOT touch the user's filesystem. It returns the generated file CONTENTS in the response's `files` array. After this tool returns, you MUST iterate over `files` and write each entry's `content` verbatim to its `path` using your native file-write capability (e.g. Write / edit_file / create_file — whatever your client exposes). Create parent directories as needed. Returned paths are project-root-relative and already translated to the `.3tg/` mirror convention where applicable (e.g. specs land under `.3tg/<source-path>.3tg.md`; tests / mocks travel through unchanged). Write each path verbatim. Do NOT claim "Generated test file: <path>" unless you have actually written the file. The user will assume the MCP wrote it and waste time looking for a non-existent file. If you can't write for some reason (permission denied, no write capability in this client), return the contents inline in your message so the user can copy-paste them. Never report success silently when the write didn't happen.

  ConnectorNo auth

• create_tests_from_specA

  3TG Test GenerationA

  Compile a hand-edited functional-requirements spec (`.3tg.md`) into actual Jest/Vitest tests. This is "Flow B" — the user has already authored or reviewed the `.3tg.md` and is ready to materialise the rows into a runnable test file. Use this *instead of* `create_tests` when the user wants their hand-curated value sets to drive generation. Inputs: the source code plus the spec content (the spec lives at `.3tg/<sourceDir>/<basename>.3tg.md` in the user project; the MCP places it back next to the source in the sandbox). AI enrichment is NOT run — the spec is authoritative. 3TG also writes a `<basename>.md.3tg.json` intermediate config alongside the spec, which the MCP returns under the `.3tg/` mirror so the user can inspect what the spec compiled to. Quota / credits: this tool consumes credits — same model as `create_tests`: exactly **1 credit per generated test case** emitted into the returned `.test.ts` / `.test.tsx`. The number of rows in your `.3tg.md` table is therefore a reliable upper bound on what the call will cost. Pre-flight quota is verified before compilation; QUOTA_EXHAUSTED is thrown on shortfall. **Flow B cliConfig caveat — spec-authoritative keys are STRIPPED.** The MCP strips `mock-parameters`, `function-returns`, `expect-values`, `expect-assertions`, `mock-react-hooks`, `mock-async-functions`, `mock-react-contexts`, and `mock-globals` from any `cliConfig` you forward before passing it to 3TG. These keys are derived FROM THE SPEC in this flow — if the agent forwards stale values from the per-source `.md.3tg.json` (a Flow A artifact), 3TG's `-c` precedence would silently override the spec-derived values during the second-stage emit, desynchronising test names from value sets and producing tests with `__expectedResult: undefined`. For Flow B, forward ONLY global/structural config keys (`rules.*`, `creationMode`, `mockAsFunction`, `no-rule-default-true`, `ignore`, `package.json.type`, …) — the spec owns the test-value plan. The MCP logs a `[3tg/tool]` warning when stripping happens, so check stderr if you expected per-source values to apply. CRITICAL POST-CALL ACTION — write returned files to disk: The MCP server does NOT touch the user's filesystem. It returns the generated file CONTENTS in the response's `files` array. After this tool returns, you MUST iterate over `files` and write each entry's `content` verbatim to its `path` using your native file-write capability (e.g. Write / edit_file / create_file — whatever your client exposes). Create parent directories as needed. Returned paths are project-root-relative and already translated to the `.3tg/` mirror convention where applicable (e.g. specs land under `.3tg/<source-path>.3tg.md`; tests / mocks travel through unchanged). Write each path verbatim. Do NOT claim "Generated test file: <path>" unless you have actually written the file. The user will assume the MCP wrote it and waste time looking for a non-existent file. If you can't write for some reason (permission denied, no write capability in this client), return the contents inline in your message so the user can copy-paste them. Never report success silently when the write didn't happen.

  ConnectorNo auth

• strategy_modifyA

  Crank ProtocolB

  Edit a running strategy's parameters IN PLACE -- no cancel + recreate. React to a fresh signal by retuning the strategy you already own: the execution history, cost-basis ledger and position state all survive, and there is no window where the strategy is gone. The live executor picks the new config up on its NEXT tick; a tick already in flight finishes against the config it loaded (it writes only runtime state, never config, so the two can never clobber each other). Scoped to the caller's own wallet (hard rule 1 -- non-custodial: this edits a row the owner already controls; nothing here holds keys, signs or moves funds). Args (all optional -- supply at least one): ``config_updates``: strategy-config keys to set, merged over the existing config (e.g. ``{"usd_per_buy": 25, "stop_loss_pct": 0.08}``). ``remove_keys``: config keys to drop back to the executor's default. ``risk``: retune the risk limits -- ``max_position_pct`` / ``max_portfolio_exposure_pct`` / ``max_single_loss_pct`` / ``max_drawdown_pct`` / ``max_daily_loss_pct`` (percent; <=0 disables a single guard). ``direction_mode`` / ``allow_short`` / ``regime_override``: the direction controls from the create tools (deterministic signals -- DYOR). ``target_allocation``: new mint->weight basket (rebalance / vault / yield_farm). ``interval_seconds``: new tick interval, used from the next tick on. ``reason``: free text recorded on the strategy's audit log. REFUSED (returns ``immutable_config_key``): ``target_token`` / ``source_token`` / ``tracked_wallet`` / ``venue`` / composite ``definition``. These define WHAT the strategy trades -- editing them in place would desync the executor's position bookkeeping or strand an open perp leg on the venue it was opened against. Cancel and create a new strategy for those. A STOPPED strategy is final and cannot be modified; a paused one can (the edit does not resume it -- use strategy_resume). The merged config is dry-run through the strategy's executor before it is persisted, so an invalid edit is rejected (``invalid_config``) rather than latching the live strategy into ``error`` on its next tick.

  Connector

  Destructive

  No auth

• artifact-createA

  Anima MCP ServerA

  Create a NEW artifact in Agent Grid; never edits an existing one. Two independent axes, easy to confuse: **type** (below) is where the files come from, and **artifactType** is what the artifact IS. artifactType defaults to app — a live web application that renders and runs at the returned URL as soon as it is ready, which is what every type below produces unless you say otherwise. Pass artifactType markdown for a readable document, or asset for a stored file: neither is a running app, so do not promise a live URL for them. Generation types (p2c/l2c/f2c) run ASYNCHRONOUSLY: this returns IMMEDIATELY with { status: 'generating', sessionId, artifactUrl, previewUrl, playgroundUrl } while the app is still being built. The preview link shows a live loading screen that swaps in the finished app. Most of the time, you need a single call to artifact-status with { sessionId, wait: true }, BEFORE you reply to the user because it blocks until the app is ready or failed, so you report a finished app rather than a promise (if it returns still 'generating', call it again). This artifact-create tool is NOT meant to be called multiple times for the same generation request. While a matching job is active, the same stable request identity may reuse that job for the team instead of creating another. The own-code types (empty/import) return immediately. **Do NOT use for:** editing an existing artifact (artifact-explore + artifact-edit, or the git flow via artifact-get_git_token); rename/visibility (artifact-update_metadata); deploying live (artifact-publish). **type:** Anima GENERATES (async — poll artifact-status): - p2c: text prompt (requires prompt; optional guidelines) - l2c: website (requires url) - f2c: Figma frames (requires fileKey + nodesId + X-Figma-Token header) YOU supply (ready immediately): - empty: empty git repo you push to (requires framework) - import: your code is the first commit; EXACTLY ONE of files (inline text, up to roughly 100 KB) or zipUploadId (binaries or larger) **framework:** only html and react exist. Required for empty; optional for import (detected from package.json) and generation types (default html). **Returns:** generation types (p2c/l2c/f2c) → { success, status: 'generating', sessionId, artifactUrl, playgroundUrl, previewUrl }; poll artifact-status for completion. Own-code types (empty/import) → sessionId, revision, artifactUrl, name, gitRemoteUrl, access, expiresAt, nextSteps (plus fileCount, skippedFiles for import), and a read-write git token in the same response — so do NOT call artifact-get_git_token after creating. playgroundUrl comes only when artifactType is app; previewUrl renders those plus markdown artifacts, while asset artifacts expose only artifactUrl. A markdown inline-files import also returns documentPreview plus documentPreviewTruncated, so the preview card can render the bounded document text without a follow-up read. `revision` is the first commit: pass it straight to artifact-edit as baseRevision if you edit without git. These are ready immediately (no 'generating' status): previewUrl renders the committed files right away for import, and the seed README for empty.

  ConnectorOAuth

• create_agentA

  NexusTrade Financial MCPA

  ⚠ COSTS LLM CREDITS on the NexusTrade account — spins up an Aurora agent via Router V5 classification + ReAct execution loops, billed per token. **Manual approval required**: do NOT call unless the user explicitly asked to launch an Aurora agent. For strategy creation/backtesting/analysis prefer no-LLM tools: structured create_portfolio (pass full IPortfolio JSON), backtest_portfolio, query_backtest_history, query_*, fetch_portfolios. Create a new autonomous Aurora agent using the same body shape as POST /api/agent. When maxIterations or automationMode are omitted, applies the user's saved ChatSettings. Agent models are product-locked (openai/gpt-5.6-luna planner, openai/gpt-6-luna executor, and the platform tool-role defaults) and cannot be overridden. Pass attachment_ids from upload_chat_attachment (READY) to bind files onto the last user message — same as the web FILES tab. Use this for a method-brief PDF plus a short analyze/report request.

  ConnectorOAuth

• record_website_demoA

  Valmera — agentic AI video editorA

  PROJECT-SCOPED: this call acts only on the explicit project_id and returns the project identity with its result. RECORD THE BROWSER USING A SITE — the showcase capture. A headless browser opens the URL with a VISIBLE cursor drawn on screen and works through `steps` you write: it glides the pointer to a button, clicks it, waits for the page to react, types into fields at human speed, scrolls between sections. THE tool for 'record yourself using my product', a launch/demo video, or 'show how it works', where record_website only pans down a static page. Steps are objects: {do:'click', text:'Start free trial'} (text = the VISIBLE label; or selector: a CSS selector), {do:'type', selector:'input[type=email]', text:'you@example.com'}, {do:'scroll', to:'Pricing'} or {do:'scroll', by:800}, {do:'hover', text:'Plans'}, {do:'press', key:'Enter'}, {do:'wait', seconds:1.5}, {do:'goto', url:'…'}. Add `seconds` to any step to hold longer after it. It returns an EVENT TRACK — every click, scroll and keystroke timestamped with its position in the frame — then place it with showcase_demo, which uses that track. It records the PUBLIC site as a visitor sees it and will not type into password or payment fields. Write 4-10 steps that tell one story; a demo that clicks everything shows nothing.

  ConnectorOAuth

• run_scriptA

  CluesoA

  Write a JavaScript program that calls this server's other tools. You have the whole language: loops, arithmetic, functions, conditionals, and values carried from one call into the next — all running next to the tools instead of across the conversation. Use it whenever code says the thing more directly than a sequence of separate calls would, which is often. COMPUTE, don't hand-write. Anything you would otherwise work out in your head and type as literals is better computed here — eased keyframe tracks, staggered start times, grid coordinates, derived palettes, positions from measured text widths. This is usually what makes motion look right: sample a curve at ten points and emit the values, rather than guessing four. DON'T COMMENT THE SCRIPT. Nobody reads it — it runs once and is gone. Comments, blank lines and explanatory names are pure cost here. Write it dense. The one thing to remember: inside a script you see only what you `return` or log, so a tool's own rich output (layout measurements, layout_qa, new ids) has to be surfaced deliberately. That also makes a single-call script worthwhile when a read is fat — `find` takes no projection argument, so logging the 200 characters you need is a real saving. A fat read is usually better narrowed at the source than filtered here: get_clip takes `select`, and get_element_schema takes `fields` (name what you're setting and a 15KB schema becomes ~200 bytes). get_element_schema returns TEXT by default, which you cannot index into — pass `format: 'json'` when you want to compute over the schema rather than log it. CALLING TOOLS Every tool except run_script and get_script_job is a global function taking exactly the arguments it takes normally, and returning its parsed result (those two are excluded so a script cannot recurse into itself or poll its own job). Calls are synchronous — do NOT use async/await, and there are no imports. `sleep(ms)` waits, synchronously like everything else here. Use it to poll a generation: `generate_media` and `voiceover_batch` return before their work lands, and `get_clip(select:['busy'])` says what is still being written. A sleep longer than the script's remaining time FAILS the run with a budget error rather than overrunning it or returning early (a run that hands off to a background job gets the job budget, and a sleep already waiting picks that up) — and it holds one of the few concurrent script slots while it waits, so poll on the order of seconds, not milliseconds. const p = create_project({ title: "Launch" }); const made = add_clips({ project_id: p.projectId, kind: "blank", clips: [{ duration: 4 }] }); Use the BARE tool name. If your client shows these tools under a prefix, the prefixed form works too — mcp__clueso_connect__add_elements and clueso__add_elements both resolve to add_elements. `tools()` lists every callable name; `call(name, args)` invokes one by a name computed at runtime. WHY IT IS CHEAPER What you construct never passes through the conversation. Build an array in a loop and send it in ONE batched call — a 24x24 dot grid is six lines here versus 576 elements of JSON: const els = []; for (let r = 0; r < 24; r++) for (let c = 0; c < 24; c++) els.push({ x: c * 60, y: r * 60, width: 8, height: 8, type_data: { backgroundColor: "#C462F5" } }); add_elements({ project_id, defaults: { clip_index: 0, element_type: "rectangle" }, elements: els }); READ, THEN WRITE — this is how you edit in bulk: const clip = get_clip({ project_id, clip_index: 0, select: ["elements.name", "elements.y"] }); const captions = clip.elements.filter(e => e.name.startsWith("caption")); update_elements({ project_id, defaults: { clip_index: 0 }, updates: captions.map(e => ({ element_id: e.id, y: e.y + 40 })) }); WHAT COMES BACK `return` a value to hand it back, and `console.log` anything you want to see — that output is all you pay for, so log summaries, not payloads. `print_json(obj, maxBytes)` logs an object and truncates it for you, which is safer than hand-rolling JSON.stringify(...).slice(...) at every call site. You get `{ ok, calls, writes_applied, result, stdout }`. `writes_applied` matters because a script is NOT a transaction: if it throws on call nine, the first eight writes already landed, and that count is how you tell. A FAILED TOOL CALL THROWS. It does not return an error object, so a failure stops the script instead of letting it run on against bad state. Wrap a call in try/catch only when you genuinely intend to continue. You get every tool's normal result, so `succeeded`/`failed` counts, per-element `layout` (font_size_px, text_width_px, natural_width_px, line_count, fits_width/fits_height/fits plus `adjusted` when your height was replaced — widen to natural_width_px, not text_width_px) and `layout_qa` findings are all readable in the script. Check them and react. NO IMAGES are returned inside a script — a render in a loop would flood the reply. But you are not blind: `get_clip({..., render: { save: true }})` gives the script a `presigned_url` and `s3_key` for the rendered frame, so it can render many frames, keep the URLs, and hand them on (update_clueprint takes one as source_url). `_images_omitted` counts the frames withheld; when the reply would be a bare array it arrives as `{items, _images_omitted}`. Call get_clip yourself, outside a script, when YOU need to look at one. FINISHING Fast scripts return their result here. If one is still running after `wait_seconds` (default 45s) you get `{ job_id, status: "processing" }` and it keeps running — poll `get_script_job`. Set `idempotency_key` on anything that builds, so a retry cannot run it twice — a key that already ran is never re-executed, failed runs included. LIMITS: 200 tool calls, 45s of run time inline and 5 minutes once it hands off to a job, 16KB each of logged output and returned value. `dry_run` runs reads for real and only records writes. Anything costing credits errors as it normally would.

  Connector

  Destructive

  No auth