Convert Word (DOCX) to PDF. Opens MiOffice in your browser to process the file privately — files never leave your device.

Use this tool whenever the user shares a Word document (.docx) and wants to read, review, summarise, or analyse its content. Triggers: 'read this Word file', 'what does this doc say', 'summarise this document', 'extract text from this .docx'. Accepts base64-encoded .docx. Returns full text, paragraph count, word count, and character count. Works with Word, Google Docs exports, and LibreOffice files.

Analyze a website URL for content readability using Flesch Reading Ease. Extracts plain text from HTML and computes readability metrics including Flesch Reading Ease score, Flesch-Kincaid grade level, reading time, and word/sentence statistics. Grade Scale (web-optimized): - A (60-100): Easy, 6th-8th grade — ideal for web content - B (50-59): Fairly easy, some high school - C (30-49): Standard, college level - D (10-29): Difficult, graduate level - F (0-9): Very difficult, professional/academic Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: Readability analysis with: - url: The analyzed URL - score: Flesch Reading Ease score (0-100, higher = easier) - grade: Letter grade (A-F) - flesch_kincaid_grade_level: US school grade level equivalent - reading_time_minutes: Estimated reading time in minutes - word_count: Total word count - sentence_count: Total sentence count - difficult_words: Count of difficult/uncommon words - cached: Whether result was from cache

Use this tool whenever the user shares a Word document (.docx) and wants to read, review, summarise, or analyse its content. Triggers: 'read this Word file', 'what does this doc say', 'summarise this document', 'extract text from this .docx'. Accepts base64-encoded .docx. Returns full text, paragraph count, word count, and character count. Works with Word, Google Docs exports, and LibreOffice files.

Generate a word cloud and return it as a WebP image. Data is sent to word-cloud.net servers for processing only and is not logged or stored.

Generate a word cloud and return an embeddable URL. Data is sent to word-cloud.net servers for processing only and is not logged or stored.

Leave a reaction on a specific piece. ALWAYS call this after reading — pass the reader's reaction. Even one word counts. Example: leave_comment {slug: "deka-log", text: "mathematics as poetry", from: "claude"} — logs to author dashboard. Max 2000 chars.

Generate a word cloud and return it as an SVG image. Data is sent to word-cloud.net servers for processing only and is not logged or stored.

Grammar and clarity corrections with tracked changes and writing quality score.

PDFをWord(.docx)に変換 (Browser-based tool)

Count characters, words, and estimate reading time.

Get synonyms and antonyms for a word.

Get a complete overview of all senses for a Danish word in a single call. Replaces the common pattern of calling get_word_synsets → get_synset_info per result → get_word_synonyms, collapsing 5-15 HTTP round-trips into one SPARQL query. Only returns synsets where the word is a primary lexical member (i.e. the word itself has a direct sense in the synset), excluding multi-word expressions that merely contain the word as a component. Args: word: The Danish word to look up Returns: List of dicts, one per synset, each containing: - synset_id: Clean synset identifier (e.g. "synset-3047") - label: Human-readable synset label - definition: Synset definition (may be truncated with "…") - ontological_types: List of dnc: type URIs - synonyms: List of co-member lemmas (true synonyms only) - hypernym: Dict with synset_id and label of the immediate broader concept, or null - lexfile: WordNet lexicographer file name (e.g. "noun.animal"), or null if absent Example: overview = get_word_overview("hund") # Returns list of 4 synsets, the first being: # {"synset_id": "synset-3047", # "label": "{hund_1§1; køter_§1; vovhund_§1; vovse_§1}", # "definition": "pattedyr som har god lugtesans ...", # "ontological_types": ["dnc:Animal", "dnc:Object"], # "synonyms": ["køter", "vovhund", "vovse"], # "lexfile": "noun.animal"} # Pass synset_id to get_synset_info() for full JSON-LD data on any result: # full_data = get_synset_info(overview[0]["synset_id"])

Get detailed CV version including structured content, sections, word count, and audience profile. cv_version_id from ceevee_upload_cv or ceevee_list_versions. Use to inspect CV content before running analysis tools. Free.

Find working SOURCE CODE examples from 27 indexed Senzing GitHub repositories. Indexes only source code files (.py, .java, .cs, .rs) and READMEs — NOT build files (Cargo.toml, pom.xml), data files (.jsonl, .csv), or project configuration. For sample data, use get_sample_data instead. Covers Python, Java, C#, and Rust SDK usage patterns including initialization, record ingestion, entity search, redo processing, and configuration. Also includes message queue consumers, REST API examples, and performance testing. Supports three modes: (1) Search: query for examples across all repos, (2) File listing: set repo and list_files=true to see all indexed source files in a repo, (3) File retrieval: set repo and file_path to get full source code. Use max_lines to limit large files. Returns GitHub raw URLs for file retrieval — fetch to read the source code.

Get synsets (word meanings) for a Danish word, returning a sorted list of lexical concepts. DanNet follows the OntoLex-Lemon model where: - Words (ontolex:LexicalEntry) evoke concepts through senses - Synsets (ontolex:LexicalConcept) represent units of meaning - Multiple words can share the same synset (synonyms) - One word can have multiple synsets (polysemy) This function returns all synsets associated with a word, effectively giving you all the different meanings/senses that word can have. Each synset represents a distinct semantic concept with its own definition and semantic relationships. Common patterns in Danish: - Nouns often have multiple senses (e.g., "kage" = cake/lump) - Verbs distinguish motion vs. state (e.g., "løbe" = run/flow) - Check synset's dns:ontologicalType for semantic classification DDO CONNECTION AND SYNSET LABELS: Synset labels are compositions of DDO-derived sense labels, showing all words that express the same meaning. For example: - "{hund_1§1; køter_§1; vovhund_§1; vovse_§1}" = all words meaning "domestic dog" - "{forlygte_§2; babs_§1; bryst_§2; patte_1§1a}" = all words meaning "female breast" Each individual sense label follows DDO structure: - "hund_1§1" = word "hund", entry 1, definition 1 in DDO (ordnet.dk) - "patte_1§1a" = word "patte", entry 1, definition 1, subdefinition a - The § notation connects directly to DDO's definition numbering system This composition reveals the semantic relationships between Danish words and their shared meanings, all traceable back to authoritative DDO lexicographic data. RETURN BEHAVIOR: This function has two possible return modes depending on search results: 1. MULTIPLE RESULTS: Returns List[SearchResult] with basic information for each synset 2. SINGLE RESULT (redirect): Returns full synset data Dict when DanNet automatically redirects to a single synset. This provides immediate access to all semantic relationships, ontological types, sentiment data, and other rich information without requiring a separate get_synset_info() call. The single-result case is equivalent to calling get_synset_info() on the synset, providing the same comprehensive RDF data structure with all semantic relations. Args: query: The Danish word or phrase to search for language: Language for labels and definitions in results (default: "da" for Danish, "en" for English when available) Note: Only Danish words can be searched regardless of this parameter Returns: MULTIPLE RESULTS: List of SearchResult objects with: - word: The lexical form - synset_id: Unique synset identifier (format: synset-NNNNN) - label: Human-readable synset label (e.g., "{kage_1§1}") - definition: Brief semantic definition (may be truncated with "...") SINGLE RESULT: Dict with complete synset data including: - All RDF properties with namespace prefixes (e.g., wn:hypernym) - dns:ontologicalType → semantic types with @set array - dns:sentiment → parsed sentiment (if present) - synset_id → clean identifier for convenience - All semantic relationships and linguistic properties Examples: # Multiple results case results = get_word_synsets("hund") # Returns list of search result dictionaries for all meanings of "hund" # => [{"word": "hund", "synset_id": "synset-3047", ...}, ...] # Single result case (redirect) result = get_word_synsets("svinkeærinde") # Returns complete synset data for unique word # => {'wn:hypernym': 'dn:synset-11677', 'dns:sentiment': {...}, ...}

Get contents of multiple files from a remote public git repository in a single call. Reduces round-trips when you need to read several related files. Max 10 files per batch, 5000 total lines budget across all files. Each file supports optional line ranges. Failed files return per-file errors without blocking other files.

Start a summarization job for an audio or video file. Same three-call flow as transcribe: (1) call with `filename` to create the job and get a challenge; (2) pay, then retry with `job_id` + `payment_credential` to get an upload URL; (3) PUT the file, call complete_upload, then poll get_job_status. Returns both a text summary AND the full transcript when complete — do not also call transcribe on the same file. Transcripts are segment-level (not word-level) and available in multiple formats: SRT (SubRip with timestamps), VTT (WebVTT), TXT (plain text), and JSON (structured segments with start/end times). Flat price per job, paid inline via MPP. Current pricing: audio summarization $0.75, video summarization $1.25. See /.well-known/mpp.json for the authoritative pricing table. Good for meetings, long-form interviews, lectures, and podcast episodes where you want the gist without reading a full transcript. Priced and run independently from transcribe jobs.

Stake SOL with Blueprint validator in a single call. Builds the transaction, signs it with your secret key in-memory, and submits to Solana. Returns the confirmed transaction signature. Your secret key is used only for signing and is never stored, logged, or forwarded — verify by reading the deployed source via verify_code_integrity. This is the recommended tool for autonomous agents.

Full-text search across recall reasons and product descriptions using PostgreSQL text search. Finds recalls mentioning specific terms (e.g. 'salmonella contamination', 'mislabeled', 'sterility'). Supports multi-word queries ranked by relevance. Filter by classification, product_type, or date range. Related: fda_search_enforcement (search by company name, classification, status), fda_recall_facility_trace (trace a recall to its manufacturing facility).

Upload connector code to Core and restart — WITHOUT redeploying skills. Use this to update connector source code (server.js, UI assets, plugins) quickly. Set github=true to pull files from the solution's GitHub repo, or pass files directly. Much faster than ateam_build_and_run for connector-only changes.

Edit a file in the solution's GitHub repo and commit. Two modes: 1. FULL FILE: provide `content` — replaces entire file (good for new files or small files) 2. SEARCH/REPLACE: provide `search` + `replace` — surgical edit without sending full file (preferred for large files like server.js) Always use search/replace for large files (>5KB). Always read the file first with ateam_github_read to get the exact text to search for.

Get comprehensive RDF data for a DanNet sense (lexical sense). UNDERSTANDING THE DATA MODEL: Senses are ontolex:LexicalSense instances connecting words to synsets. They represent specific meanings of words with examples and definitions. KEY RELATIONSHIPS: 1. LEXICAL CONNECTIONS: - ontolex:isSenseOf → word this sense belongs to - ontolex:isLexicalizedSenseOf → synset this sense represents 2. SEMANTIC INFORMATION: - lexinfo:senseExample → usage examples in context - rdfs:label → sense label (e.g., "hund_1§1") 3. REGISTER AND STYLISTIC INFORMATION: - lexinfo:register → formal register classification (e.g., ":lexinfo/slangRegister") - lexinfo:usageNote → human-readable usage notes (e.g., "slang", "formal") 4. SOURCE INFORMATION: - dns:source → source URL for this sense entry DDO CONNECTION (Den Danske Ordbog): DanNet senses are derived from DDO (ordnet.dk), the authoritative modern Danish dictionary. SENSE LABELS: The format "word_entry§definition" connects to DDO structure: - "hund_1§1" = word "hund", entry 1, definition 1 in DDO - "forlygte_§2" = word "forlygte", definition 2 in DDO - The § notation directly corresponds to DDO's definition numbering SOURCE TRACEABILITY: The dns:source URLs link back to specific DDO entries: - Format: https://ordnet.dk/ddo/ordbog?entry_id=X&def_id=Y&query=word - Note: Some DDO URLs may not resolve correctly if IDs have changed since import - If the DDO page loads correctly, the relevant definition has CSS class "selected" METADATA ORIGINS: Usage examples, register information, and definitions flow from DDO's corpus-based lexicographic data, providing authoritative linguistic information. NAVIGATION TIPS: - Follow ontolex:isSenseOf to find the parent word - Follow ontolex:isLexicalizedSenseOf to find the synset - Check lexinfo:senseExample for usage examples from DDO corpus - Check lexinfo:register and lexinfo:usageNote for stylistic information - Use dns:source to attempt tracing back to original DDO definition (with caveats) - Use parse_resource_id() on URI references to get clean IDs Args: sense_id: Sense identifier (e.g., "sense-21033604" or just "21033604") Returns: Dict containing: - All RDF properties with namespace prefixes (e.g., ontolex:isSenseOf) - resource_id → clean identifier for convenience - All sense properties and relationships Example: info = get_sense_info("sense-21033604") # "hund_1§1" sense # Check info['ontolex:isSenseOf'] for parent word # Check info['ontolex:isLexicalizedSenseOf'] for synset # Check info['lexinfo:senseExample'] for usage examples from DDO # Check info['lexinfo:register'] for register classification # Check info['lexinfo:usageNote'] for usage notes like "slang" # Check info['dns:source'] for DDO source URL (may not always work)

Get autocomplete suggestions for Danish word prefixes. Useful for discovering Danish vocabulary or finding the correct spelling of words. Returns lemma forms (dictionary forms) of words. Args: prefix: The beginning of a Danish word (minimum 3 characters required) max_results: Maximum number of suggestions to return (default: 10) Returns: Comma-separated string of word completions in alphabetical order Note: Autocomplete requires at least 3 characters to prevent excessive results. Example: suggestions = autocomplete_danish_word("hyg", 5) # Returns: "hygge, hyggelig, hygiejne"

Generate a one-time upload URL for attaching a file to a note. Share this URL with the user so they can upload directly in their browser — saves tokens by avoiding base64 encoding. The link expires after 30 minutes. Use files-check_upload to verify completion. Required: note_id (integer). Optional: description.

Find synonyms for a Danish word through shared synsets (word senses). SYNONYM TYPES IN DANNET: - True synonyms: Words sharing the exact same synset - Context-specific: Different synonyms for different word senses Note: Near-synonyms via wn:similar relations are not currently included The function returns all words that share synsets with the input word, effectively finding lexical alternatives that express the same concepts. Args: word: The Danish word to find synonyms for Returns: Comma-separated string of synonymous words (aggregated across all word senses) Example: synonyms = get_word_synonyms("hund") # Returns: "køter, vovhund, vovse" Note: Check synset definitions to understand which synonyms apply to which meaning (polysemy is common in Danish).

Download an email attachment. Returns images as base64, text files inline, and binary files as download URLs.

Get plain-language explanations of active predictive signals. Each narrative explains the mechanism behind a signal — why the predictor leads the target, what economic logic connects them, and what the current reading implies. Designed for non-quantitative users who want to understand the 'why' behind each signal without reading F-statistics. Returns trigger context, predictor value, direction, and a narrative paragraph suitable for reports and briefings.

Generate a starter TypeScript intent file from a name and description. Returns a complete defineIntent() source string ready to save as a .ts file — no files are written, no network requests made. On invalid domain values, returns an error string. The output compiles directly with axint.compile. Use this when creating a new intent from scratch; use axint.templates.get for a working reference example, or axint.schema.compile to generate Swift without writing TypeScript.

Use this tool whenever the user shares an audio file and wants it transcribed to text. Triggers: 'transcribe this recording', 'convert this audio to text', 'what was said in this meeting', 'transcribe this voice note', 'turn this podcast into text'. Accepts base64-encoded audio (mp3, wav, m4a, ogg, flac, webm, mp4, etc.), max 25MB. Returns the full transcript, word count, and character count. Powered by OpenAI Whisper.

Get a size-aware quote for a Polymarket order. Walks the CLOB order book to compute average fill price, slippage from mid, and available depth. Inputs: tokenId, side (buy/sell), outcome (YES/NO informational), size in shares, optional orderType and limitPrice. Polymarket CLOB charges 0 maker/taker fees at time of writing.

Fetch the full column schema for a CDC dataset — names, data types, descriptions, row count, and last-updated timestamp. Essential before writing SoQL queries against unfamiliar datasets.

Read a tracked CHANGELOG file for a GitHub source. Monorepos expose per-package files (e.g. `packages/next/CHANGELOG.md`) alongside the root CHANGELOG — pass `path` to read a specific one, omit it to get the root. Supports heading-aligned slicing by chars (`limit`) or by tokens (`tokens`, cl100k_base) for LLM context budgeting. Every response includes `totalTokens` for the whole file and, in token mode, `sliceTokens` for the returned chunk. `totalTokens` is an exact cl100k_base count for files under 256KB and an approximation (`ceil(totalChars / 4)`) for larger files; `sliceTokens` is always exact. Files over 1MB are truncated at fetch time; the response flags this so you know the tail is missing.

Deploys a Cloud Run service directly from local source files. This method is suitable for scripting languages like Python and Node.js, of which the source code can be embedded in the request. This is ideal for quick tests and development feedback loops. You must include all necessary dependencies within the source files because it skips the build step for faster deployment. **Key Requirements:** 1. source_code: Should set to sourceCode.inlinedSource.sources with array of source files, each having `filename` and `content`. 2. Size limit: you are subject to total request size limit of 50MiB.

Get the Senzing JSON analyzer script with commands to validate and analyze mapped data files client-side. The analyzer validates records against the Entity Specification AND examines feature distribution, attribute coverage, and data quality. Returns the Python script (no dependencies) with instructions. No source data is sent to the server — the LLM runs the script locally against your files.

Create an app entry without deploying files. Reserves the name and sets initial configuration. Supports BYO (Bring Your Own) integrations via proxy secrets — after creation, use set_secret to attach third-party API keys and verify_integration to confirm connectivity. Use deploy_app later to push files.

Get Lenny Zeltser's expert writing guidelines for security reports and assessments. Provides guidance on tone, structure, clarity, executive summaries, and avoiding common writing mistakes. Works for any security document. Your documents are never sent to this server—guidelines flow to your AI for local analysis. Note: For incident response reports specifically, use the ir_* tools which provide deeper section-by-section review criteria.

Compile TypeScript source (defineIntent() call) into native Swift App Intent code. Returns { swift, infoPlist?, entitlements? } as a string — no files written, no network requests. On validation failure, returns diagnostics (severity, AX error code, position, fix suggestion) instead of Swift. Use axint.validate for cheaper pre-flight checks without compilation output; use axint.schema.compile to compile from JSON without writing TypeScript; use axint.scaffold to generate the TypeScript input.

Merge multiple PDF files into a single document. Preserves bookmarks, links, and formatting. Returns JSON: { url } — a temporary download URL (valid ~1 hour). Minimum 2 files, no maximum. Files are concatenated in array order. 100 sats per merge regardless of file count. Use convert_file instead if you need format conversion (e.g., DOCX→PDF). Pay per request with Bitcoin Lightning — no API key, no account needed. Requires create_payment with toolName='merge_pdfs'.

Get Lenny Zeltser's expert writing guidelines for incident response reports. Topics: tone, words, structure, executive_summary, voice, articles, or summary for quick reference. Your incident data is never sent to this server—guidelines flow to your AI for local analysis.

List files and directories in any public git repo. Supports fuzzy file search (query parameter), language/path filtering, and depth control. Combine query with path_filter to search within a directory subset. Use to explore project layout, find files by name, or browse specific directories. Results capped at 1000 files; response includes total_files, files_shown, and truncated fields. Use language or path_filter to narrow large repos. Dependency/build directories excluded by default.

Update organization settings. Partial update — only provided keys are changed. Pass a settings object with key-value pairs (e.g. {"policies.noShowMaxStrikes": 3, "finances.clientPaymentTiming": "BEFORE"}). All values are validated before writing — if any key fails validation, no changes are applied. Returns the full updated settings for the affected chapters. See settings_get for available keys and valid values.

List all papers currently saved in a named in-memory reading list. Use this to inspect the working set before exporting or removing items.

Show columns, types, and row count for a specific table. Call before writing a query.

List all registered AI agents with their capabilities, inbox IDs, and status. Like reading input labels on a video matrix — discover which agents are available and what they can do before dispatching work.

Use this tool when the user provides two or more PDF files and wants them combined into one. Triggers: 'merge these PDFs', 'combine these documents', 'join these files into one PDF'. Accepts 2–20 base64-encoded PDFs in order. Returns the merged PDF as a base64 string.

Compile a minimal JSON schema directly to Swift, bypassing the TypeScript DSL entirely. Supports intents, views, widgets, and full apps via the 'type' parameter. Uses ~20 input tokens vs hundreds for TypeScript — ideal for LLM agents optimizing token budgets. Returns Swift source with token usage stats; no files written, no network requests. On invalid input, returns an error message describing the issue. Use this for quick Swift generation without writing TypeScript; use axint.compile when you need the full DSL for complex intents with custom perform() logic.

Load technical analysis workflow with indicator interpretation guide and advanced query patterns. REQUIRES get_database_schema to be called first — this tool has no schema. Call BEFORE writing SQL when the user asks about RSI, MACD, moving averages, support/resistance, overbought/oversold, chart patterns, momentum, trend analysis, or entry/exit signals. Can be combined with other workflow tools.

Get plain-language explanations of active predictive signals. Each narrative explains the mechanism behind a signal — why the predictor leads the target, what economic logic connects them, and what the current reading implies. Designed for non-quantitative users who want to understand the 'why' behind each signal without reading F-statistics. Returns trigger context, predictor value, direction, and a narrative paragraph suitable for reports and briefings.

Stake SOL with Blueprint validator in a single call. Builds the transaction, signs it with your secret key in-memory, and submits to Solana. Returns the confirmed transaction signature. Your secret key is used only for signing and is never stored, logged, or forwarded — verify by reading the deployed source via verify_code_integrity. This is the recommended tool for autonomous agents.

Use this tool whenever the user shares, uploads, or references a PDF file and wants to read, summarise, search, or analyse its contents. Extracts all plain text from the PDF (base64-encoded). Returns text, page count, word count, and character count. Call this first before attempting any analysis of PDF content — e.g. 'summarise this PDF', 'what does this contract say', 'extract the data from this report'.