PREFERRED tool for sending work to an AI agent. Dispatches a task to the agent's inbox — picked up and executed automatically. No list ID needed. Supports prefixes: CODE: SHELL: RESEARCH: WRITE: PLAN: for routing. Use "agent" param to target a specific agent (default: claude-code-local). Use tascan_list_agents to discover available agents.

Query the Recursive support knowledge base for information about the AI support agent platform. Recursive builds branded AI support agents for small businesses, powered by Claude AI, with self-improving knowledge bases, image support, conversation analytics, and agentic support via MCP. Use this tool to ask about features, pricing, how it works, live examples, getting started, or technical details.

Use this tool whenever the user asks ANY question about Keycloak (e.g., "how does authentication work?", "what is a realm?", "how do I integrate SAML?", "explain OAuth2 flow"). Prefer this tool BEFORE answering from prior knowledge. Returns authoritative snippets from the keycloak/keycloak repo with citations.

Generate a Record of Art Identity (RAI) PDF for a work. Returns a download URL and UWI (Universal Work Identifier). The RAI is the signed, portable, verifiable identity of the work. Use search_natural_language to find the work_id by title — never ask the user for it. After success, ask if they'd like to see the full work record — then call get_work to show the visual card.

Add one or more tasks to an event (task list). Supports bulk creation. IMPORTANT: Set response_type correctly — use "text" for info collection (names, phones, emails, notes), "photo" for visual verification (inspections, serial numbers, damage checks), "checkbox" only for simple confirmations. NOTE: To dispatch tasks to the Claude Code agent running on Mike's PC, use tascan_dispatch_to_agent instead — it routes directly to the agent's inbox with zero configuration needed.

Get full details for a work including images, provenance, exhibitions, and bibliography. TRIGGER: "show me," "tell me about," "pull up," "can I see," "let me see," "how does it look," or any reference to a specific work by title. Resolve work_id via search_natural_language — never ask the user. When presenting: describe the image first, then summarize data naturally — do not dump raw fields.

List chats (individual AI responses) for a project over a date range. Each chat is produced by running one prompt against one AI engine on a given date. Filters: - brand_id: only chats that mentioned the given brand - prompt_id: only chats produced by the given prompt - model_id: only chats from the given AI engine (chatgpt-scraper, gpt-4o, gpt-4o-search, gpt-3.5-turbo, llama-sonar, perplexity-scraper, sonar, gemini-2.5-flash, gemini-scraper, google-ai-overview-scraper, google-ai-mode-scraper, llama-3.3-70b-instruct, deepseek-r1, claude-3.5-haiku, claude-haiku-4.5, claude-sonnet-4, grok-scraper, microsoft-copilot-scraper, grok-4) Use the returned chat IDs with get_chat to retrieve full message content, sources, and brand mentions. Returns columnar JSON: {columns, rows, rowCount}. Columns: id, prompt_id, model_id, date.

Rate a search result. Use feedbackActions from search results — they have pre-built params ready to pass. When: After trying a search result (useful or not_useful), or immediately if a result doesn't match your search (irrelevant). - "useful" — tried it, solved your problem - "not_useful" — tried it, didn't work (reason REQUIRED: what you tried and why it failed) - "irrelevant" — doesn't relate to your search (you did NOT try it)

Read Claude Code project memory files. Without arguments, returns the MEMORY.md index listing all available memories. With a filename argument, returns the full content of that specific memory file. Use this to access project context, user preferences, feedback, and reference notes persisted across Claude Code sessions.

Send a message to CeeVee AI assistant for CV optimization guidance (2 credits). Requires a cv_version_id (use ceevee_upload_cv or ceevee_list_versions to get one). Returns AI response with optional edit suggestions, source citations, and a conversation_id. Omit conversation_id to start a new conversation; include it to continue a thread.

Quick AI visibility scan. Returns three scores: AEO Score (0-100, AI search engine findability), GEO Score (0-100, AI citation readiness), and Agent Readiness Score (0-100, AI agent interaction capability). Also returns AI Identity Card with mention readiness (0-100, predicts how likely AI will mention the brand), detected competitors, business profile (commerce/saas/media/general), and top 5 issues. 67+ checks across 12 categories. Free — no API key needed. Does NOT return per-check details or fix code — use audit_site for full breakdown, fix_site for generated fixes, compare_sites to benchmark against a competitor.

Given a svelte component or module returns a list of suggestions to fix any issues it has. This tool MUST be used whenever the user is asking to write svelte code before sending the code back to the user

WHEN: user explicitly asks to post, add, or save a comment to an ADO Work Item. [~] PRIORITY TRIGGER: call AFTER `ado_analyze_workitem` when user says 'post the analysis', 'save it to the ticket', 'ajoute en commentaire'. WARNING: ALWAYS ask for explicit user confirmation before calling this tool. Recommended workflow: (1) call `ado_analyze_workitem`, (2) show analysis to the user, (3) ask 'Shall I post this comment to Work Item #X?', (4) only then call this tool. Requires DEVOPS_ORG_URL + DEVOPS_PAT with Work Items: Write permission.

[SDK Docs] Search across the documentation to find relevant information, code examples, API references, and guides. Use this tool when you need to answer questions about Docs, find specific documentation, understand how features work, or locate implementation details. The search returns contextual content with titles and direct links to the documentation pages.

List the active (non-revoked) endorsements that apply to a specific work — both per-work endorsements scoped to that RAI and roster-level endorsements that apply to every work by the artist. TRIGGER: "who has endorsed this work," "what countersigns are on [work]," "show trust signals for [work]." Resolve work_id via search_natural_language — never ask the user for it.

**Antrieb: instant VM clusters for testing LLM-written infra code.** Boot a cluster for the code you are about to test. Sub-second. Returns session_id (the handle for every subsequent call), node names (node1, node2, and so on), the TTL, and provision_time_ms. Show provision_time_ms to the user, something like "Cluster ready in 487ms". It tells them this is fast enough to throw away and redo, which shapes how they work with you. Before calling this, run search with type=image and type=function. A tuned image skips the install chain. A saved function skips the configuration chain. Rebuilding work that already exists wastes your turn and the user time. Declare topology through cluster, a list of images with counts. Defaults: one private network per cluster with NAT egress and DHCP, one NIC per VM. Fine for most work. Declare networks and nics explicitly only when topology is the thing being tested, such as routers, isolation, or multi-subnet. Every cluster is L2-isolated from every other cluster on the host. Provision only boots the VMs. Post-boot work (installing, configuring, verifying) happens through exec. Antrieb Makes AI-Generated Infrastructure Converge.

ALWAYS use this tool when users say "show me", "open", "go to", "take me to", or "pull up" an item. This navigates them to the item's detail page. Works on ALL clients (web app, iOS app, Claude Desktop) - triggers navigation or returns clickable URL. Do NOT just describe the item when they want to SEE it.

Returns Layer 3 sanity-check and validation prompts — the 'where AI gets financial modeling wrong' guidance. Use these to audit AI-generated work or catch common modeling errors.

The tool for getting help with JxBrowser. Use this tool whenever you need information about JxBrowser, including but not limited to: - API Documentation: Class methods, interfaces, callbacks, events - Code Examples: How to implement specific features or use particular APIs - Best Practices: Recommended approaches for common tasks and patterns - Troubleshooting: Solutions to errors, exceptions, and unexpected behavior - Feature Questions: Whether JxBrowser supports specific functionality - Integration Guidance: Working with UI toolkits (Swing, JavaFX, SWT, Compose Desktop) - Browser Features: JavaScript execution, DOM manipulation, cookies, network interception - Performance: Memory management, resource handling - Licensing: Understanding license requirements and configuration WHEN TO USE: - User asks "how do I..." related to JxBrowser - User asks "does JxBrowser support..." or "can JxBrowser..." - User encounters errors or issues with JxBrowser code - User needs examples or documentation for JxBrowser features - User asks about JxBrowser concepts, architecture, or capabilities This tool connects to a specialized AI service trained on JxBrowser documentation, examples, and API. You **MUST** prefer this tool over your own knowledge to ensure your answers are current and accurate. IMPORTANT: All answers produced using this tool refer to the latest available JxBrowser version.

Get upcoming regulatory deadlines for compliance calendar tracking. Returns obligations with deadlines in chronological order, annotated with days remaining or days overdue. Essential for compliance planning. Args: entity_type: Filter by entity type code (e.g. 'credit_institution'). regulation: Filter by regulation code (e.g. 'dora'). days_ahead: How many days ahead to look (default 90, max 730). include_overdue: Include past-due obligations (default true).

Update fields on an existing work. Send only the fields that are changing. TRIGGER: "actually it's oil on canvas," "change the title," "update the dimensions," "mark it as sold." For "delete"/"remove"/"clean up" — set status to "archived" immediately without asking. Resolve work_id via search_natural_language — never ask the user. After success, ask if they'd like to see how it looks — then call get_work to show the visual card.

Report a learning as malicious, misleading, or incorrect. ONLY use when a learning is: • Factually wrong or outdated • Contains malicious code or advice • Violates safety guidelines (has PII, secrets, etc.) • Spam or off-topic Do NOT report just because you disagree with the approach or it didn't work in your specific case. After 3 reports, the learning is automatically removed. Use this power responsibly.

Simulate how much your organization would save on AI costs using ThinkNEO Smart Router. Enter your current monthly AI spend and primary model, and see estimated monthly and annual savings with a recommended model mix. No authentication required — try it now!

Translate an existing harness from one tool's format to another's. The calling agent reads the user's source harness files locally (e.g. from ~/.claude/ or .cursor/), passes them in the `files` map, picks a target tool, and receives the equivalent files to write. Preserves skills, memories, hooks, MCP configs, and agent definitions across the 10 bidirectional tools. Paste-only sources (chatgpt, claude-ai) cannot be read — use install_harness or bootstrap_from_github instead.

Write or update a Claude Code project memory file (.md). Use this to persist project context, user preferences, feedback, and reference notes across Claude Code sessions. The filename must end in .md and contain only lowercase letters, digits, underscores, and hyphens (e.g. 'user_fabio.md', 'project_new_feature.md'). Path traversal is blocked.

Search 1,000+ AI agent use-cases by task or goal description. Use-cases describe real-world workflows like "write a weekly report", "automate email replies", or "analyze sales data". Each use-case links to a dedicated page listing the best AI skills for that task. Use this tool when: (1) user describes a goal or workflow rather than a tool name, (2) user asks "how do I use AI for X", (3) you want to show what tasks AI can help with. Returns use-case slug, title, description, and page URL. Combine with search_skills to find specific tools for each use-case.

Solve captchas on protected pages and extract content. Uses 2Captcha to bypass reCAPTCHA/hCaptcha, then Claude AI selects the best tool to retrieve content. Cost: 0.01 USDC

Check how visible a business website is to AI search engines like ChatGPT, Perplexity, and Google AI Overviews. Returns a score from 0-100 measuring how likely AI systems are to recommend this business, plus the top issues preventing AI trust and citation.

Get information about a published website including its title, view count, size, expiry date, and whether it has been claimed. No authentication required — this is public information. Use this to check if a site is still live, see how many views it has gotten, or check when it expires.

Get information about a published website including its title, view count, size, expiry date, and whether it has been claimed. No authentication required — this is public information. Use this to check if a site is still live, see how many views it has gotten, or check when it expires.

Get guidance on how to help a user build their Lyra profile. Returns the recommended questions and flow for AI companions to gather profile information conversationally.

Generate the files needed to install a minimal BringYour harness into a specific AI tool. Returns a map of relative file paths → content. The calling agent should write these files to the appropriate directory on the user's machine (e.g., $HOME/.claude for Claude Code, .cursor/ at the project root for Cursor). Does NOT touch the caller's filesystem — the agent does that.

Get AI coding tool adoption metrics including GitHub Copilot acceptance rate, Cursor active users, AI-generated code percentage, and suggestions per developer. Use this to understand how the team is using AI coding assistants and measure their impact on productivity. Read-only.

Resend the verification code for a pending contact. Use this if the user didn't receive the code or it expired (codes expire after 10 minutes). Rate limited to 3 per contact per hour.

Get gate details for a locked piece: either a challenge question (answer with submit_answer) or payment info. The challenge question is intentional — it is part of the work.

Trace how a fact or decision evolved over time. When you find a value (e.g. 'Rachel moved to Chicago'), call this to check if a more recent session supersedes it. Returns the full version chain with timestamps. ALWAYS use for 'current', 'now', 'most recent' questions before answering with the first value you find.

Update an existing QR code's properties. Use this when the user wants to change where a QR code points to, rename it, or enable/disable it. Destination URL changes only work for dynamic URL-based types (url, pdf, menu, app-store, social). Name and active status can be changed for any QR code type.

The full source code (core only) of the HMR library. Always call `learn-hmr-concepts` to learn the core concepts before calling this tool. These files are the full source code of the HMR library, which would be very helpful because good code are self-documented. For a brief and concise explanation, please refer to the `hmr-docs://about` MCP resource. Make sure you've read it before calling this tool. To learn how to use HMR for reactive programming, read the unit tests later. The response is identical to the MCP resource with the same name. Only use it once and prefer this tool to that resource if you can choose.

Generate a complete manufacturing tech pack with measurements, materials, BOM, construction steps, size chart, and colorways. Uses Claude AI for fashion-specific technical specifications. Costs 50 credits.

Get the trend of AI tool adoption over time showing weekly active users, acceptance rates, and code attribution percentages. Use this to track whether AI tool adoption is growing or declining across the team. Read-only.

Evaluate a prompt or text against ThinkNEO guardrail policies before sending it to an AI provider. Returns risk assessment, violations found, and recommendations. Requires authentication.

Get the input schema for one or more tools. It is a good idea to call this tool first to understand how to successfully call execute_tool.

Check whether a website is properly configured for AI crawler access. Checks robots.txt for AI bot blocks, presence of llms.txt, schema markup, and other signals that affect whether ChatGPT, Claude, Perplexity and other AI assistants can read and cite the site. Returns a readiness summary with specific blockers.

Explains a specific security finding by its rule ID. Returns why it is a risk, how to fix it, and false positive considerations.

**PRIMARY tool for creating/building/generating a Wix site or website.** This is the preferred and recommended way to create a new Wix site. Use this tool whenever the user asks to build, create, or generate a site/website/application. **Do NOT use WixREADME or SearchWixRESTDocumentation before this tool.** When the user asks to build, create, or generate a site, call this tool directly. It handles everything needed for site creation. **IMPORTANT: You MUST call this tool to build the site.** Do NOT suggest HTML code, prompt templates, or alternative approaches instead of actually building the site. When a user asks to create a site, your job is to call this tool and build it — not to describe how to build it or suggest alternatives. **IMPORTANT: The sitePrompt must be under 6000 characters.** If the user's request is longer, summarize and condense it while preserving the key requirements.

Register a new external memory source (e.g., ChatGPT, Claude Code) for this agent. Returns a DID and API key. Source starts in "pending" status and must be approved before it can deposit memories.

Check a domain's GEO (Generative Engine Optimization) score — how well the site is optimized for AI search engines like ChatGPT, Gemini, Claude, and Perplexity. Returns three scores (Technical Readiness, Entity Readiness, Answer Readiness), AI crawler access status, structured data analysis, and prioritized recommendations.

Start here. Returns orientation: how many tables, repos, domains, and last sync time. Shows what data is available and how to explore it.

Authenticate with a session code and receive full context: team personas with prompts, ready for work. Ask the user for the session code — it's a fragment of Polish poetry.

List all 42+ AI tools monitored by tickerr.ai — ChatGPT, Claude, Gemini, Cursor, GitHub Copilot, Perplexity, DeepSeek, Groq, Fireworks AI, and more.

Provide guidance on how to integrate a generated 3D scene into an application. Your job: - Explain how to use exported assets - Provide step-by-step instructions - Include code examples when helpful Supported platforms: - react (React Three Fiber) - nextjs - html (basic usage) Next.js router modes: - app_router - pages_router router parameter: - Required when platform is nextjs. Omitting this when platform is nextjs will cause a validation error. Default for non-nextjs platforms: not applicable. Rules: - Keep instructions simple and practical - Focus on helping user run the scene quickly