"How to Find and Fix Bugs in a React Native App" matching MCP tools:

• get_starter_templateA

  NativeExpressA

  The free MIT React Native starter repo: its stack, what it already does, what it deliberately leaves out, and the commands to get it running. Use this when someone is about to start a mobile app project, or asks what to build an Expo or React Native app on top of.

  ConnectorNo auth

• deploy_appA

  DockholdA

  Deploy a GitHub repository as a live web app on Dockhold. Call this when the user wants to put an app online, get a shareable HTTPS URL, or host a demo. Returns the new app id. Two paths: a PUBLIC repo needs only repo_url; a PRIVATE repo needs repo_url plus github_installation_id (call list_github_repos first, each repo comes with the installation_id to pass here). Deploying a private repo turns on auto-deploy: future pushes to that repo redeploy the app automatically. The app builds and comes online automatically; poll get_app_status to watch it. Set memory_mb to size the app's compute, one of the values get_resource_usage reports under compute.steps_mb: 256 MB fits a static site or a small API, 512 MB fits a typical Node or Python web app, and anything that holds data in memory needs more. Omit it and the app gets the minimum slice (256 MB) so it doesn't take your whole compute pool; resize_app changes the size later with no rebuild, applied as a rolling update that replaces the app's instances. The response reports memory_mb (what this app got) and compute_available_mb (what's left in your pool), so size the next app off that. This tool needs a GitHub repo URL: if the code only exists locally (no repo), it cannot be used here, and the user should run `npx dockhold login` then `npx dockhold deploy` in the project folder instead. Requires a token with the deploy scope.

  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

• patternfetch_analogsA

  patternfetchA

  Find earlier windows IN THE SAME SERIES whose shape resembles the current price action and return the FULL distribution of what followed (win-rate, median, min, max, n) over a fixed forward horizon. Parameters: window = how many recent bars form the shape being matched (default 32); horizon = how many bars forward each match is measured over (default 20). WHEN: an agent wants the historical spread of outcomes after a similar-looking setup, including how wide and how uncertain that spread is. WHEN NOT: you want the current technical picture (use brief), you want to find candidates across the market (use scan), or you need one expected value — this deliberately returns a distribution, not a point estimate. NOT a prediction, NOT a backtest of a strategy; past distribution does not guarantee future results. Example: {"ticker":"ETH/USDT","timeframe":"1d"}. Impersonal data, not advice.

  ConnectorNo auth

• leave_feedbackA

  AgentMarketplaceA

  Leave feedback about AgentMarketplace itself — bugs, confusing tools, feature requests. This is for feedback about the platform, not about a counterparty. There is no reputation or rating system here — that lives in AgentTrust. Args: access_token: AgentAuth bearer token (requires ``market.read``). message: Free-text feedback, up to 2000 characters. Returns: ``{"recorded": true}`` on success.

  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

• adsap_google_create_app_campaignA

  AdsapA

  Create a complete Google App campaign in ONE atomic request: budget, campaign (ALWAYS created PAUSED — activate later with adsap_google_set_entity_status), location + language targeting, an ad group, and the app ad. App campaigns promote a mobile app across Google Search, Google Play, YouTube, Discover and the Display Network — Google mixes the ad from your text (and optional media) automatically. The ad links to the app's store listing AUTOMATICALLY: there is no URL input, ever. Any PUBLIC store app id works (no app ownership or Play/Firebase link is required to create): Android app_id is the package name (e.g. com.example.app), iOS app_id is the numeric App Store id (e.g. 570060128) with app_store 'apple_app_store'. v1 optimizes INSTALLS at your target_cpi (target cost per install, plain number in the account currency); Google recommends a daily_budget of roughly 50x target_cpi — lower is allowed but slows learning (a warning says so). Text: 2-5 headlines (max 30 chars) + 1-5 descriptions (max 90). Optional media: image_asset_ids / video_asset_ids from adsap_google_upload_asset (images 1.91:1, 4:5 or 1:1 up to 5MB; videos are YouTube assets 10-60s) — text-only is fully valid. In-app-action, value, engagement and pre-registration goals are NOT supported in v1. If the account has no conversion tracking a warning is returned — relay it (Android installs normally auto-track once a Google Play link exists). Use dry_run: true first — Google validates the whole chain without creating anything; show the plan, get confirmation, then execute. ad_account_id is the bare 10-digit customer ID. For website campaigns use adsap_google_create_search_campaign / adsap_google_create_pmax_campaign; for Meta app ads use adsap_create_campaign with an app objective.

  ConnectorAPI key

• update_password_policyA

  The BridgeA

  Set the session token lifetimes on the Bridge app: accessTokenTTL and refreshTokenTTL, both in SECONDS. Only the fields you pass are changed. IMPORTANT — despite the name (kept for parity with `bridge auth password-policy`), this sets token TTLs and NOTHING about passwords: Bridge exposes no password-complexity, rotation or reuse policy on the management API, so if the user asked for password rules, say so instead of calling this. The access token TTL controls how often a client silently refreshes (short is safer, e.g. 900 = 15 minutes); the refresh token TTL controls how long a user stays logged in without re-authenticating (e.g. 2592000 = 30 days). Read the current values with get_auth_config. Before writing app code that consumes this, call get_integration_guide with topic=auth and the framework the app uses.

  ConnectorOAuth

• edit_contentA

  AthenaHQB

  Apply exact find-and-replace edits to a content draft's article body. Deterministic, no AI model involved: your replacement text lands verbatim (use revise_content when you want an AI rewrite from an instruction). Each edit's find text must match the current body exactly once; zero or multiple matches fail the whole call and nothing is saved, so extend the find text until it is unique. Edits apply in order, later ones see earlier results, and the batch is recorded as one new version, recoverable via get_content_versions / restore_content_version. Pass expected_version_number from get_content_draft to fail with a conflict when a new version was recorded after your read (a revision, edit, or restore). An app autosave changes the body without recording a version; that drift is caught by the anchors themselves, since a moved or vanished find text fails the call.

  ConnectorNo auth

• sixta_explain_errorA

  sixta-connectA

  Decode a database error and get the fix and the next step — no connection needed. Paste a MySQL error number (1213, 1062, 1452, 1205…) or a PostgreSQL SQLSTATE (40P01, 23505, 53300…), optionally with the failing statement, and get the proximate cause, the concrete fix, and — when it helps — the SIXTA tool and artifact to go deeper (e.g. a deadlock → paste SHOW ENGINE INNODB STATUS for sixta_explain_deadlock). Use when the user pastes a DB error code or message. Input is analyzed in memory and never stored.

  ConnectorOAuth

• sixta_rewrite_queryA

  sixta-connectA

  Turn a flagged anti-pattern into the safe, equivalent rewrite — no connection needed. Paste a SQL query and get ready-to-run rewrites anchored to deterministic rules: `= NULL` → `IS NULL`, `NOT IN (subquery)` → `NOT EXISTS` (NULL-safe), deep OFFSET → keyset pagination, `ORDER BY RAND()` → a keyed random sample — each with its semantics caveat spelled out. Every literal rewrite is then re-analyzed in-process and reported as 'clean' or 'still flags X', so the safe rewrite is self-checked — no need to feed it back through sixta_analyze_query. Use when the user asks 'how do I fix / rewrite this query' or after sixta_analyze_query flags a smell. Input is analyzed in memory and never stored.

  ConnectorOAuth

• get_resource_usageA

  DockholdA

  Show your account's compute, database-RAM, and storage pools: how much you've bought, how much is used, and how much is free, plus every app's current size. Call this before any resize tool (the allowed sizes come from its steps fields), and to explain to the user why an app ran out of memory or a deploy was refused for capacity.

  ConnectorNo auth

• acc_list_projectsA

  ACC MCPA

  Enumerate every ACC and BIM 360 project the authenticated APS app can see by walking all accessible hubs and their project lists. When to use: The agent needs to discover project IDs before calling any other tool (e.g. the user says 'show me my projects' or 'find issues in the Tower project' and no project_id is known yet). Also useful to confirm hub membership for a project. When NOT to use: Do not call this repeatedly in a loop — cache the result; if the user already supplied a project_id starting with 'b.', skip discovery. APS scopes: data:read account:read. No write scope needed. Rate limits: APS default ~50 req/min per app per endpoint; BIM 360 hubs endpoints are pageable (limit 200). This tool fans out 1 hubs call + N project calls (one per hub) so call it sparingly on tenants with many hubs. Errors: 401 (APS token expired — refresh and retry once); 403 (app not provisioned in the BIM 360/ACC account — ask user to have an account admin add the APS client_id); 404 (rare, indicates hub deleted mid-call); 429 (rate limit — back off 60s); 5xx (ACC upstream — retry with jitter). Side effects: None. Read-only and idempotent.

  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

• forgetA

  memxusA

  Permanently delete one memory by UUID. When to use: user asks to remove outdated or incorrect context, or to free plan storage. When NOT: fix content → update (mode=replace); find the ID first → list_memories or recall. Requires delete OAuth scope. Non-idempotent: deleting the same memory_id twice fails. Errors: Memory not found, Not authorized to delete this memory. Side effects: removes the memory row and vector embedding with no recovery; invalidates plan cache. The target workspace is always the one the memory itself belongs to (echoed in resolved_workspace); optionally pass workspace: <name> as a safety confirmation — the call fails if the memory is not actually in that workspace.

  Connector

  Destructive

  No auth

• how_munimji_helpsA

  HelloBooks AI AgentsA

  Explain how HelloBooks and Munimji (the in-app AI assistant) help a specific business — given a free-text description of the user's own operations. Returns a curated capability knowledge base: business-operation areas (sales, purchases, banking, tax, reports, inventory, payroll, multi-entity, setup), and for each AI capability WHO does the work — `autonomous` (Munimji does it on its own, e.g. OCR extraction, running reports), `approval` (Munimji prepares the entry and you one-click approve before it posts to the ledger, e.g. AI categorization, find-and-match, creating invoices/bills by chat), `assist` (co-pilot, e.g. guided onboarding, voice), or `manual` (a software feature you run yourself). Each capability links to the backing software features. Use this when a user describes their business and asks "how can HelloBooks help me?", "what can the AI do for my shop/practice/agency?", or "what can Munimji do on its own vs what do I approve?". Pass their description in `businessDescription`; optionally filter by `area` or `autonomy`. The AI never posts to a ledger without approval. For the full software catalog call list_features; for pricing call list_plans.

  ConnectorNo auth

• how_munimji_helpsA

  HelloBooks AI Agents MCP ServerB

  Explain how HelloBooks and Munimji (the in-app AI assistant) help a specific business — given a free-text description of the user's own operations. Returns a curated capability knowledge base: business-operation areas (sales, purchases, banking, tax, reports, inventory, payroll, multi-entity, setup), and for each AI capability WHO does the work — `autonomous` (Munimji does it on its own, e.g. OCR extraction, running reports), `approval` (Munimji prepares the entry and you one-click approve before it posts to the ledger, e.g. AI categorization, find-and-match, creating invoices/bills by chat), `assist` (co-pilot, e.g. guided onboarding, voice), or `manual` (a software feature you run yourself). Each capability links to the backing software features. Use this when a user describes their business and asks "how can HelloBooks help me?", "what can the AI do for my shop/practice/agency?", or "what can Munimji do on its own vs what do I approve?". Pass their description in `businessDescription`; optionally filter by `area` or `autonomy`. The AI never posts to a ledger without approval. For the full software catalog call list_features; for pricing call list_plans.

  ConnectorNo auth

• feedbackA

  HomespunA

  Reports a problem with homespun itself to the relay operator, and lists what this agent has already reported. A report is the operator's only visibility into a failure that happened inside an agent's session, so an unreported one is a failure nobody can fix. The channel covers homespun's own behaviour: a 5xx, or an error code the guide does not describe; a disagreement between documented and observed behaviour; something the tool surface cannot express, such as a missing capability or a schema that contradicts itself; an app misbehaving in a way that traces back to the platform (the bridge, the runtime, serving, the data API) rather than to authored HTML; or a guide that was wrong, ambiguous or silent. Outside its scope: the human's own task; bugs in an app the agent authored; presentation preferences, which belong in `taste`; the human's own configuration, such as a missing API key or the wrong account; and a 4xx caused by the agent's own arguments, except where the error message itself was misleading, which is a documentation problem best filed as a `note`. Duplicates cost the operator triage rather than adding signal. Action `list` returns this agent's own submissions, newest first, so a failure already recorded needs no second row: one report covers one distinct failure, however many times it was retried. The operator sees the row and not the session, so a bare "deploy failed" is not actionable. An actionable `message` carries the surface (mcp, cli, relay or app-runtime); where it happened (the tool or route); the skill version, from the `<!-- homespun skill vX.Y.Z -->` comment at the top of the guide; what was expected, in one line; what was observed, in one line carrying the exact error code and message; and the minimal steps or arguments that reproduce it. `type` is bug for something broken, feature for something missing, note for a rough edge or a confusing doc. `app_id` scopes a report to one app. There is no reply channel, so a report is not a route to an answer. Actions: create files one report; list returns this agent's own submissions, newest first, paginated by `before`.

  ConnectorNo auth

• list_rankingsA

  Game Dev Outsourcing DirectoryA

  List every ranked list the directory publishes — "Overall", "3D Art in Poland", "Unreal Engine" and so on — with each list's size, URL and slug, plus the "method" string describing exactly what the order measures. Use to find the right list before calling get_ranking. Always pass the method on: these lists are ordered by size, years in business and how completely a listing is filled in, not by studio quality, and are not an endorsement.

  ConnectorNo auth

• get_app_guide_sectionA

  Wellness ProjectA

  Look up customer-facing Wellness Project product knowledge. Use it for app navigation, visible features, integrations, metric meanings, logging methods, permissions/sync questions, or troubleshooting. Wellness Project-specific answers are closed-book: answer them from this tool's result, never from memory. The result states its own grounding rules. For how-to questions, select the relevant feature topic; a matching published video link is included when one exists. For a request to find, report through, or open the feedback form, use topic=troubleshooting. In-app chat must offer the open_feedback action chip, not print the link (chat links render as plain text). External MCP clients may provide the exact Share feedback link returned. Never claim the form was opened without the user tapping the action. For recognized exercise names or whether an exercise is in the app's exercise library, use list_exercises rather than guessing or expecting the app guide to enumerate exercises. For cost, price, Free vs Pro, Founding Member, upgrading, or the 3-analysis limit, use topic=pricing. Reproduce exactly one returned pricing message verbatim, including its subscription link when it has one, and add no other pricing detail. The app attaches its own upgrade button to that message when one applies; never type out a button, chip, link markup, or call-to-action of your own.

  ConnectorOAuth