// MCP TOOL REFERENCE

CROW TOOLS

576 tools from the Crow MCP Server, categorised by risk level.

View the Crow policy →
// ALL 576 CROW TOOLS
READ 279 tools
Read dozzle_container_url Build a Dozzle URL that deep-links straight to a container Read adguard_query_log Recent DNS queries. Returns the last N entries with timestamp, client, question, response, and whether it w... Read adguard_stats Query stats: queries, blocked counts, top clients, top domains, top blocked domains (defaults to last 24h w... Read adguard_status AdGuard Home status: version, protection state, upstream DNS servers, filter count, stats summary Read caddy_cert_health Report TLS cert health across all configured sites: ok / warning / error per domain, with expiry, ACME issu... Read caddy_list_sites List all sites currently declared in the Caddyfile Read caddy_status Report Caddy status: admin-API reachability, number of configured sites, and TLS/ACME summary Read changedetection_get_watch Get full detail on a single watch, including configuration and last-check metadata. Read changedetection_list_changes List recent change events across all watches. Useful for Read changedetection_list_watches List all configured watches (URLs being tracked) with their last-check and last-change timestamps. Read crow_actual_accounts List all Actual Budget accounts with current balances Read crow_actual_budgets Get budget categories and amounts for a given month Read crow_actual_get_transaction Get detailed information about a specific transaction by ID Read crow_actual_reports Generate a spending summary report. Groups transactions by category, payee, or month. Read crow_actual_transactions List transactions with optional filters. Amounts in cents (negative = expense, positive = income). Read crow_audiobookshelf_browse Browse a library Read crow_audiobookshelf_collections List collections and series in a library Read crow_audiobookshelf_get_item Get detailed information about an audiobook or podcast (chapters, duration, narrators, progress) Read crow_audiobookshelf_libraries List all Audiobookshelf libraries (audiobook and podcast collections) Read crow_audiobookshelf_play Get a stream URL for an audiobook or podcast episode. Returns a playback URL. Read crow_audiobookshelf_progress Get all items currently in progress (listening progress across all libraries) Read crow_audiobookshelf_search Search audiobooks and podcasts across all libraries by title, author, or narrator Read crow_audit_log Return recent audit log entries for a project space, ordered newest first. Read crow_backend_schema Show discovered tools and schema for a data backend. Read crow_blog_stats Get blog statistics Read crow_bookstack_books List books in BookStack, optionally filtered or sorted Read crow_bookstack_chapters List chapters in a BookStack book Read crow_bookstack_get_page Get a page Read crow_bookstack_search Full-text search across all BookStack content (pages, chapters, books, shelves) Read crow_bookstack_shelves List all shelves in BookStack with book counts Read crow_browser_discover_selectors Dump all interactive elements on the page (inputs, buttons, links, selects) with their attributes. Read crow_browser_extract_links Extract all links from the current page with their text and URLs. Read crow_browser_extract_pdf Extract text (or metadata) from a PDF by URL or local path, using pdfjs-dist. Good for gov reports, papers,... Read crow_browser_extract_tables Extract HTML tables from the current page as structured JSON data. Read crow_browser_extract_text Extract clean article text from the current page using Mozilla Readability. Strips ads, nav, and boilerplat... Read crow_browser_load_session Restore a previously saved session (cookies and storage). Read crow_browser_paginate Follow pagination links and collect content from multiple pages. Returns combined results. Read crow_browser_scrape Extract structured data from the page using CSS selectors. Define a schema of selector → field name mappings. Read crow_browser_screenshot Take a screenshot of the current page or a specific element. Read crow_browser_status Check browser container and CDP connection health. Read crow_calibre_browse_category Browse books within a specific category (e.g., a particular author, tag, or series) Read crow_calibre_download Get a download URL for a book in a specified format (EPUB, PDF, MOBI, AZW3, etc.) Read crow_calibre_get_book Get detailed metadata for a specific book by ID Read crow_calibre_list_books List books in the Calibre library with sorting and pagination Read crow_calibre_list_categories List available categories in the Calibre library (authors, tags, series, publishers, languages, ratings) Read crow_calibre_search Search the Calibre library by title, author, or keyword. Returns matching books with metadata. Read crow_calibreweb_download Get a download URL for a book in a specified format, or get the web reader URL Read crow_calibreweb_get_book Get detailed information about a specific book by ID, including reader URL and download links Read crow_calibreweb_list_books List books in the Calibre-Web library. Browse by newest, oldest, or rated. Read crow_calibreweb_search Search the Calibre-Web library by title, author, or keyword. Returns matching books. Read crow_calibreweb_shelves List available shelves (bookshelves/collections) in Calibre-Web Read crow_campaign_crawl_subreddit Crawl a subreddit Read crow_campaign_get_subreddit Retrieve cached subreddit intelligence from the DB (rules, flairs, timing, culture notes). Read crow_campaign_list List campaigns with summary stats. Optionally filter by status. Read crow_check_notifications Check pending notifications. Returns unread notifications filtered by type. Automatically cleans expired an... Read crow_consulting_get Get a single consulting pipeline row by tea_id. Read crow_consulting_list_by_stage List consulting pipeline rows at a given stage. Read crow_consulting_list_pending List organizations at stage= Read crow_consulting_stats Count consulting pipeline rows per stage. Read crow_data_list_databases List all SQLite databases registered as data backends. Optionally filter by project. Read crow_data_list_saved List saved queries, charts, and case studies. Read crow_data_preview Quick preview of a table Read crow_data_query Execute a read-only SQL query on a user-owned SQLite database. Only SELECT, EXPLAIN, PRAGMA, and WITH are a... Read crow_data_schema Get the schema (tables, columns, types, row counts, indexes) for a registered SQLite database. Read crow_deep_recall Proactive recall: search ALL knowledge sources (memories, research sources, research notes, blog posts) for... Read crow_discover Discover available actions and their parameter schemas. No args = list categories. category = list actions.... Read crow_discover_relays List configured relays and discover new ones. Relays enable offline message delivery and contact discovery. Read crow_dream Analyze memory health: find stale, contradictory, or redundant memories. Returns a consolidation report wit... Read crow_find_contacts Search for Crow users by email hash. Privacy-preserving: only SHA-256 hashes are compared, never plain text... Read crow_frigate_clip_url MP4 clip URL for a Frigate event. Event must have has_clip=true (check via list_events). Read crow_frigate_latest_by_label Most recent Frigate event for a label, across all cameras (unless a camera is given). Read crow_frigate_list_cameras List all configured Frigate cameras with detect/record state, resolution, and tracked objects. Reads /api/c... Read crow_frigate_list_events List Frigate detection events with optional filters. Cached for 30s at the HTTP layer so chatty polling doe... Read crow_frigate_snapshot Get a snapshot URL. Either Read crow_frigate_stats Frigate system stats: version, detector inference time, CPU usage per process. Cached for 30s. Read crow_get_chord_diagram Get an SVG chord diagram for a chord name Read crow_get_context Generate and return the full crow.md cross-platform behavioral context document. This document defines how ... Read crow_get_file_url Get a presigned download URL for a file Read crow_get_memory Fetch a single memory by ID. Returns a compact JSON blob with id, content, category, importance, tags, and ... Read crow_get_post Get full blog post by ID or slug Read crow_get_setlist Get setlist details with songs Read crow_get_source Get full details of a specific source by ID. Read crow_gis_geocode Forward geocode: convert an address or place name to latitude/longitude coordinates. Uses self-hosted Nomin... Read crow_gis_reverse Reverse geocode: convert latitude/longitude to an address. Zoom level controls detail (3=country, 10=city, ... Read crow_gis_search Search for places by name or type within an optional bounding box. Returns locations with coordinates. Read crow_glasses_search_photos Search the glasses photo library by caption or extracted text (OCR). Returns top matches with presigned URL... Read crow_glasses_status List paired Meta Ray-Ban Meta (Gen 2) glasses devices and their connection state. Read crow_identity_list List attestations for this Crow instance. Includes both active and revoked entries; filter with include_rev... Read crow_identity_verify Verify an attestation for a given (crow_id, app, handle) triple. Fetches the latest non-revoked attestation... Read crow_inbox Check your inbox for received shares and messages from contacts. Read crow_iptv_channels Browse IPTV channels with optional filters. Returns current EPG info when available. Read crow_iptv_epg Get EPG (Electronic Program Guide) for a channel. Read crow_iptv_list_playlists List all IPTV playlists with channel counts. Read crow_iptv_stream_url Get the stream URL for a channel. Read crow_jellyfin_browse Browse Jellyfin library by collection or media type with sorting and pagination Read crow_jellyfin_collections List all Jellyfin libraries and virtual folders (Movies, TV Shows, Music, etc.) Read crow_jellyfin_get_item Get detailed information about a specific Jellyfin library item Read crow_jellyfin_now_playing Get active Jellyfin sessions with current playback information Read crow_jellyfin_play Get a direct stream URL for a Jellyfin item. Use for playback or sharing. Read crow_jellyfin_search Search the Jellyfin media library by title/name. Returns matching items with metadata. Read crow_kavita_browse Browse Kavita series in a library with pagination Read crow_kavita_get_series Get detailed information about a Kavita series including volumes, chapters, and metadata Read crow_kavita_libraries List all Kavita libraries with type and folder info Read crow_kavita_reading_progress Get reading progress for a Kavita series Read crow_kavita_recently_added List recently added series in Kavita Read crow_kavita_search Search Kavita for manga, comics, and ebooks by title or keyword Read crow_kb_list_articles List knowledge base articles with optional filtering. Read crow_kb_review_flags List resources that have been flagged as potentially outdated, awaiting human review. Read crow_kb_search Full-text search across knowledge base articles. Read crow_kodi_browse Browse Kodi library by type with sorting and pagination Read crow_kodi_now_playing Get detailed info about what is currently playing: title, progress, elapsed/total time, speed Read crow_kodi_search Search Kodi library by title/name. Returns matching items with basic metadata. Read crow_kodi_status Get Kodi system info: version, active players, and status Read crow_lifecycle_snapshot Return the current orchestrator lifecycle snapshot (refcounts per provider, last released timestamps). Read crow_linkding_get Get detailed information about a specific Linkding bookmark Read crow_linkding_list List Linkding bookmarks with optional tag filtering and pagination Read crow_linkding_search Search Linkding bookmarks by text. Searches across URL, title, description, notes, and tags. Read crow_list_backends List registered data backends with their connection status. Read crow_list_backgrounds List previously generated background images in the gallery Read crow_list_contacts List all connected peers with their online/offline status, last seen time, and sharing activity. Read crow_list_context_sections List all crow.md sections with their metadata (key, title, sort order, enabled status, device/project scope... Read crow_list_crosspost_transforms List the available (source, target) transform pairs for crow_crosspost. Each pair is a pure function in ser... Read crow_list_crossposts List recent cross-posts from the log with their status (queued/ready/published/cancelled/error). Read crow_list_files List stored files with optional filtering Read crow_list_instances List all registered Crow instances (local and remote). Shows instance name, hostname, directory, status, an... Read crow_list_members List active members of a project space with their roles and resolved capabilities. Read crow_list_memories List memories with optional filtering by category, tags, or importance. Good for browsing what Read crow_list_message_groups List all message groups with their members. Read crow_list_pipelines List available pipelines with their descriptions and default schedules. Read crow_list_posts List blog posts with filtering Read crow_list_presets List available multi-agent team presets with their descriptions, provider, model, and agent names. Read crow_list_projects List all projects with their status, type, and source counts. Read crow_list_remote_tools List tools available on remote Crow instances. Shows connected instances and their exposed tools. Read crow_list_schedules List all scheduled tasks, optionally filtering to only enabled schedules. Read crow_list_setlists List setlists Read crow_list_songs List songs in the songbook Read crow_list_sources List sources with filtering options. Read crow_list_sync_conflicts List sync conflicts between instances. When the same data is modified on multiple instances, conflicts are ... Read crow_media_briefing Generate a news briefing: AI narration script from top articles, optionally with TTS audio. Read crow_media_digest_preview Preview what a digest email would contain (top unread articles). Read crow_media_feed Browse articles from your news feed. Supports chronological or personalized Read crow_media_get_article Get full article content by ID Read crow_media_list_sources List subscribed news sources with status Read crow_media_search Full-text search across articles. Set discover_sources=true to also search the web for new RSS feeds when l... Read crow_media_stats Get media statistics overview Read crow_memory_stats Get statistics about stored memories - counts by category, total count, importance distribution. Read crow_miniflux_entries List feed entries with filters (unread, starred, by feed or category) Read crow_miniflux_feeds List all RSS feeds with unread counts and metadata Read crow_miniflux_get_entry Get the full content of a specific feed entry Read crow_navidrome_albums Browse Navidrome albums with sorting and pagination Read crow_navidrome_artists List all artists in the Navidrome library Read crow_navidrome_get_album Get album details with full track listing from Navidrome Read crow_navidrome_now_playing Show currently playing tracks across all Navidrome clients Read crow_navidrome_search Search Navidrome for songs, albums, and artists Read crow_navidrome_stream Get a stream URL for a song in Navidrome. The URL includes authentication and can be used for direct playback. Read crow_orchestrate_status Check status / retrieve the result of an orchestration job by its job id. Returns running, completed (with ... Read crow_paperless_download Get a download URL for the original or archived version of a document Read crow_paperless_get Get detailed metadata and content preview for a specific document Read crow_paperless_list List documents with optional filters (tags, correspondent, document type, date range) Read crow_paperless_search Full-text search across all documents in Paperless-ngx (searches OCR content, titles, tags, correspondents) Read crow_plex_browse Browse a Plex library section with sorting and pagination Read crow_plex_get_item Get detailed metadata for a specific Plex library item Read crow_plex_libraries List all Plex library sections (Movies, TV Shows, Music, etc.) Read crow_plex_on_deck Get the On Deck list — items ready to continue watching or recently added unwatched items Read crow_plex_search Search the Plex media library by title/name. Returns matching items with metadata. Read crow_project_get Get full details for a project space: metadata, members (with capabilities), attached backends, source/note... Read crow_project_stats Get statistics about the project database. Read crow_recall_by_context Retrieve memories relevant to a given context. Uses full-text search across content, context, and tags. Mem... Read crow_search_memories Search persistent memory using full-text search (and semantic search when available). Returns memories rank... Read crow_search_notes Search notes by content. Read crow_search_sources Full-text search across all sources. Read crow_server_status Show which servers are active/inactive and their tool counts. Read crow_sharing_status Show your Crow identity, connected peers, relay status, and sharing statistics. Read crow_shiori_get Get detailed information about a specific Shiori bookmark, including cached content status Read crow_shiori_list List Shiori bookmarks with pagination Read crow_shiori_search Search Shiori bookmarks by keyword. Searches across URL, title, and excerpt. Read crow_storage_stats Get storage usage statistics Read crow_tax_filing_guide Generate step-by-step filing instructions for IRS Free File Fillable Forms. Read crow_tax_get_documents List all uploaded and confirmed tax documents. Returns extracted data from W-2s, 1099s, 1098s that have bee... Read crow_tax_get_form Get line-by-line values for a specific form. Read crow_tax_ingest_document Read a tax document PDF (W-2, 1099, 1098-T) and extract data. Returns extracted values with confidence scor... Read crow_tax_validate Check the return for errors and warnings. Read crow_transpose_song Transpose a song to a different key (non-destructive — returns transposed content without modifying the post) Read crow_trilium_browse_tree Browse the TriliumNext note tree from a starting note, showing children to a given depth. Read crow_trilium_get_attributes Get all attributes (labels and relations) of a TriliumNext note. Read crow_trilium_get_note Get a note Read crow_trilium_recent List recently modified notes in TriliumNext. Read crow_trilium_search Full-text search across TriliumNext notes. Returns matching notes with metadata. Read crow_vikunja_get_task Get detailed information about a specific Vikunja task Read crow_vikunja_labels List all available labels in Vikunja Read crow_vikunja_projects List all Vikunja projects Read crow_vikunja_tasks List tasks in a Vikunja project with optional filters Read crow_wallabag_get Get the full content and metadata of a saved article Read crow_wallabag_list List saved articles with optional filters (archive status, starred, tags, sorting) Read crow_wallabag_search Search saved articles in Wallabag by keyword. Searches titles and content. Read crow_workspace_dir Returns the absolute filesystem workspace path for a project space. Bots and agents call this before writin... Read crowdsec_alerts Recent CrowdSec alerts. Each alert represents one detected attack pattern (triggered scenario). Read crowdsec_decisions Active decisions (bans currently enforced by any bouncer listening on the LAPI). Read crowdsec_status LAPI reachability, active decisions (bans in force), and alert counts Read dozzle_status Check whether the Dozzle log viewer is running and reachable Read forgejo_list_issues List issues in a repository Read forgejo_list_repos List repositories the current token owner can access (paginated) Read forgejo_repo_info Get detailed info for one repository Read fw_get_playlist_tracks List the tracks in a playlist by id, in playlist order. Each entry has a 0-based Read fw_list_library List the authenticated user Read fw_now_playing Most recent listening activity for the authenticated user (last N listens). Read fw_playlists List the authenticated user Read fw_search Search the local + cached federated catalog. Default scope Read fw_status Report Funkwhale pod status: reachability, version, federation mode, instance policy counts, auth whoami. Read get_dataset Retrieve a specific dataset by ID Read gitea_list_issues List issues in a repository Read gitea_list_repos List repositories the current token owner can access (paginated) Read gitea_repo_info Get detailed info for one repository Read gts_feed Fetch a timeline. Choices: home (authenticated user Read gts_review_reports List pending moderation reports (local + federated). Read-only. Read gts_search Search accounts / hashtags / statuses across the fediverse. Remote queries resolve via WebFinger. Rate-limi... Read gts_status Report GoToSocial instance health: reachability, admin account status, federation peer count, pending notif... Read immich_get_album Get album details including asset list Read immich_get_photo Get detailed metadata for a specific photo Read immich_list_albums List all photo albums Read immich_search_photos Search photos by text query, date range, or location Read lemmy_feed Fetch posts. type_: Subscribed (follows), Local (this instance), All (federated). Rate-limited: 60/hour. Read lemmy_list_communities List communities. Default scope Read lemmy_review_reports List open post + comment reports (admin / mod only). Read-only summary. Read lemmy_search Search across communities, posts, comments, users. Rate-limited: 60/hour. Read lemmy_status Report Lemmy instance health: site info, federation mode, admin list, user stats, open registrations. Read list_sources List available government data sources Read maker_get_learner Get one learner Read maker_get_session_context Return non-PII context the companion Read maker_hint Request a scaffolded hint for the current activity. Output is filtered (reading-level / blocklist / length)... Read maker_list_learners List all learner profiles. Admin-only. Read maker_next_suggestion Return a suggested next activity based on recent progress. No-op with a friendly canned reply for guest ses... Read maker_validate_lesson Validate a lesson JSON against the schema. Returns specific errors so custom lesson authors (teachers/paren... Read mastodon_feed Fetch a timeline. home = follows; public = local+federated; local = this instance; notifications = mentions... Read mastodon_review_reports List pending moderation reports (admin-only). Read-only summary. Read mastodon_search Search accounts/hashtags/statuses. Remote queries resolve via WebFinger when resolve=true. Rate-limited: 60... Read mastodon_status Report Mastodon instance health: reachability, version, user/post/domain stats, admin account, federation p... Read matrix_federation_health Ask the public Matrix Federation Tester (federationtester.matrix.org) whether this server federates correct... Read matrix_joined_rooms List rooms the authenticated user is in, with a name hint for each. Read matrix_room_messages Paginated message history for a room. Returns the most recent N events by default. Read matrix_status Report Dendrite status: reachability, server version, whoami, federation mode (disabled/enabled), and the c... Read matrix_sync One-shot sync with a short server-side timeout. Returns new events, invites, and joined-room deltas since Read netdata_alarms List Netdata alarms. By default returns only raised (warning/critical) alarms. Read netdata_charts List active Netdata charts. Use the returned chart IDs with netdata_query to fetch data points. Optionally ... Read netdata_query Query a single chart Read netdata_status Netdata agent status: version, hostname, uptime, chart count, raised-alarm count Read pf_feed Fetch a timeline. home = follows; public = local+federated; local = this instance; notifications = mentions... Read pf_review_reports List pending moderation reports (admin-only). Read pf_search Search accounts / hashtags / statuses. Remote queries resolve via WebFinger. Rate-limited: 60/hour. Read pf_status Report Pixelfed instance health: reachability, version, stats, federation peer count, authenticated account. Read pt_list_channels List the authenticated user Read pt_list_videos List videos. scope: local (this instance), federated (all), subscriptions (my follows). Rate-limited: 60/hour. Read pt_review_reports List open moderation reports (admin/moderator role). Read pt_search Search videos + channels across local + cached federated content. Rate-limited: 60/hour. Read pt_status Report PeerTube instance health: reachability, version, stats, federation peer count, transcoding config, s... Read s4_echo_pattern Echo a date code back. Call with code in YYYY-MM-DD form. Read s4_echo_plain Echo a code back. Call with any string code. Read search_datasets Search for datasets across government sources Read searxng_list_engines List the search engines configured in this SearXNG instance Read searxng_search Run a metasearch query via SearXNG and return the top results. SearXNG aggregates answers from multiple eng... Read searxng_status Check whether SearXNG is reachable and responding to healthz Read stirling_status Check whether the Stirling PDF container is reachable. Returns the HTTP status of the home page and the con... Read stirling_web_url Return the Stirling PDF web UI URL for the user to open in a browser. Stirling Read uptimekuma_metrics Fetch Uptime Kuma Read uptimekuma_status Check whether Uptime Kuma is reachable on the configured URL. Returns the HTTP status of the entry page and... Read vaultwarden_backup_info Report the size and modification time of the Vaultwarden data directory (for backup freshness checks). Does... Read vaultwarden_status Check whether Vaultwarden is reachable and return its build info Read vaultwarden_user_count Return the number of registered Vaultwarden users. Requires a valid VAULTWARDEN_ADMIN_TOKEN — the admin API... Read wf_get_post Fetch a single post Read wf_list_collections List blog collections owned by the authenticated user. Read wf_list_posts List posts in a collection (public endpoint — no auth required). Paginated. Read wf_status Report WriteFreely instance health: reachability, federation status, authenticated user, collection count. Read crow_export_blog Export published posts as Hugo or Jekyll-compatible markdown files Read crow_generate_bibliography Generate a formatted bibliography in APA, MLA, Chicago, or web citation format. Read crow_media_listen Generate or retrieve TTS audio for an article. Requires node-edge-tts package (npm install node-edge-tts). Read crow_project_capabilities Return the effective capability set for a member of a project space. Defaults to the local user; pass conta... Read crow_trilium_export Export a TriliumNext note in HTML or Markdown format. Read maker_export_learner Export all data for a learner as a JSON bundle (for parental-request responses and right-to-be-forgotten pr... Read wf_export_posts Export all posts owned by the authenticated user as JSON. Useful for migration and backup.
WRITE 187 tools
Write crow_crosspost Cross-post a status from one federated bundle to another via the shared transform library. Requires idempot... Write crow_iptv_favorite Add or remove a channel from favorites. Write crow_media_playlist_items Manage playlist items: add, remove, reorder, or list. Write crow_remove_member Soft-revoke a member from a project space (sets revoked_at; does not delete the row). Requires manage_membe... Write crow_unpublish_post Revert a published post to draft status Write wf_unpublish_post Move a collection post back to drafts (removes it from the public blog and federation). The post is NOT del... Write crow_glasses_start_note_session Begin a note-taking session. Mode selection is NOT optional — pick based on user intent: Write crow_share Share a memory, research project, source, or note with a connected contact. The data is encrypted end-to-en... Write crow_tax_new_return Create a new tax return for a given year and filing status. Write crow_blog_customize_theme Apply custom CSS overrides to the blog theme Write crow_calibreweb_reading_status Get or set reading status for a book (read, unread, reading) Write crow_data_case_study_create Create a case study with text, chart, or map sections. Case studies can be published to the Crow blog. Write crow_data_case_study_publish Publish a case study as a Crow blog post. Emits <figure> HTML with static PNG fallbacks for charts and maps... Write crow_data_case_study_rerender_figures Re-render PNG fallbacks for a case study Write crow_data_create_database Create a new empty SQLite database and register it as a project data backend. Write crow_data_save_query Save a named query or chart configuration for reuse. Write crow_dismiss_notification Dismiss a notification or snooze it for later. Snoozing hides the notification until the specified time. Write crow_gis_batch_geocode Geocode a column of addresses in a data_backend SQLite table. Adds lat/lng columns to the table with result... Write crow_glasses_confirm_action_items Confirm which action items from a just-summarized session to convert into Crow notifications. Write crow_kavita_want_to_read Manage the Kavita want-to-read list: add, remove, or list series Write crow_miniflux_star Toggle bookmark/star on a feed entry Write crow_store_memory Store a new piece of information in persistent memory. Use this whenever you learn something important abou... Write crow_verify_source Mark a source as verified or unverified. Write fw_follow Follow a library (by UUID) or a remote channel (by actor URL/handle @user@server). Rate-limited: 30/hour. Write fw_mute_user Mute a user (silence notifications + hide from feeds but keep federation). Inline; rate-limited: 5/hour. Write gts_block_user Block an account system-wide (the authenticated user no longer sees their posts and vice versa). Rate-limit... Write gts_follow Follow an account by handle (@user@domain) or local account ID. Rate-limited: 30/hour. Write gts_mute_user Mute an account (hide posts but still federate). Rate-limited: 5/hour. Write gts_unfollow Unfollow an account. Write lemmy_block_community Block a community (hide all its posts from your feeds). Rate-limited: 5/hour. Write lemmy_block_user Block a person (hide their posts + comments from your view). Rate-limited: 5/hour. Write lemmy_follow_community Subscribe to a community (local or federated). Accepts numeric community ID or Write lemmy_unfollow_community Unsubscribe from a community. Write maker_end_session Gracefully end a session. Transitions active→ending with a 5s flush window, writes wrap-up memory for non-g... Write maker_log_progress Log a lesson-progress event for the session Write mastodon_block_domain Block a remote domain at the user level (hide all accounts from that domain for the authenticated user). Fo... Write mastodon_block_user Block an account (hide their posts + block DMs from them). Rate-limited: 5/hour. Write mastodon_follow Follow an account by handle (@user@domain) or local account ID. Rate-limited: 30/hour. Write mastodon_mute_user Mute an account (hide posts but still federate). Rate-limited: 5/hour. Write mastodon_report_remote File a moderation report to a remote server about an account. Rate-limited: 5/hour. Write mastodon_unfollow Unfollow an account. Write matrix_create_room Create a new room. Rate-limited: 10/hour. Write matrix_invite_user Invite a user to a room. Rate-limited: 10/hour. Write matrix_register_appservice Register a Matrix appservice (used by F.12 matrix-bridges bundle). Writes the registration YAML into Dendrite Write pf_block_user Block an account system-wide (the authenticated user no longer sees their posts). Rate-limited: 5/hour. Write pf_follow Follow an account by handle (@user@domain) or local account ID. Rate-limited: 30/hour. Write pf_mute_user Mute an account (hide posts but still federate). Rate-limited: 5/hour. Write pf_report_remote File a moderation report to a remote server about an account. Rate-limited: 5/hour. Write pf_unfollow Unfollow an account. Write pt_block_user Block an account (hide their videos + comments). Rate-limited: 5/hour. Write pt_rate_video Like / dislike / unrate a video by numeric id. Rate-limited: 60/hour. Write pt_report_remote File a moderation report. Can report a video (video_id) or a comment (comment_id) or a whole account. Rate-... Write pt_subscribe Subscribe to a channel by handle (name@host). Rate-limited: 30/hour. Write pt_unsubscribe Unsubscribe from a channel. Write caddy_add_federation_site Add a federation-aware reverse-proxy site block. Emits directives for the chosen profile (matrix | activity... Write caddy_add_matrix_federation_port Add a :8448 site block with its own Let Write caddy_add_site Add a reverse-proxy site block to the Caddyfile and reload. For anything beyond Write caddy_set_wellknown Add or replace a standalone /.well-known/ handler at a domain that does NOT otherwise proxy to the federate... Write changedetection_create_watch Add a new URL to watch. Change Detection will begin polling it on its default schedule. Write crow_accept_invite Accept an invite code from another Crow user. This establishes a peer connection and enables sharing. Shows... Write crow_actual_update_budget Set the budget amount for a category in a given month. Amount in cents. Write crow_add_context_section Add a new custom section to crow.md. Custom sections extend Crow Write crow_add_member Add or update a member of a project space. Upsert: if the contact already has an active membership, the rol... Write crow_add_note Add a note - can be attached to a project, a source, or both. Write crow_add_relay Add a Nostr or peer relay to your configuration. Nostr relays handle encrypted messaging. Peer relays handl... Write crow_add_source Add a source with full metadata and automatic citation generation (APA, MLA, Chicago, web). Write crow_add_to_setlist Add a song to a setlist Write crow_blog_settings Get or update blog settings (title, tagline, author, theme, podcast config) Write crow_bookstack_create_page Create a new page in BookStack within a book or chapter Write crow_bookstack_update_page Update a page Write crow_browser_export Export previously scraped data as CSV or JSON file. Write crow_browser_save_session Save cookies and local storage to a file for later restoration. Write crow_calibreweb_add_to_shelf Add a book to a shelf in Calibre-Web Write crow_campaign_approve_posts Batch approve posts by ID. Moves them from pending_approval/draft to approved. Write crow_campaign_create Create a new social media campaign. Write crow_campaign_draft_post Create a draft post for a campaign targeting a specific subreddit. Write crow_campaign_generate_posts AI-generate tailored post drafts for a campaign across multiple subreddits. Uses the configured BYOAI provi... Write crow_campaign_schedule_post Schedule a post for future publishing. Sets scheduled_at and moves to appropriate status. Write crow_campaign_set_credentials Store encrypted Reddit credentials. Validates connection before saving. Write crow_campaign_update Update a campaign Write crow_campaign_update_post Update a post Write crow_consulting_promote Promote an organization to a consulting pipeline stage. Upserts the row if the tea_id is not already in the... Write crow_consulting_write_prospectus Write a finished prospectus markdown to the MPA prospectus inbox and mark the pipeline row as generated. Th... Write crow_create_message_group Create a message group for group conversations. Add contacts by name or Crow ID. Write crow_create_notification Create a notification for the user. Use for reminders, alerts, or any information the user should see in th... Write crow_create_post Create a new blog post draft Write crow_create_project Create a new project to organize sources, notes, and data backends under. Write crow_create_schedule Create a scheduled or recurring task. Uses cron expressions for timing (e.g. Write crow_create_setlist Create a setlist of songs Write crow_create_song Create a song in the songbook (stored as a blog post tagged Write crow_crosspost_mark_published Mark a queued cross-post as published (called by the target bundle Write crow_edit_post Update an existing blog post Write crow_generate_background Generate a new AI background image via SDXL Turbo. The image is set as the companion Write crow_generate_invite Generate a single-use invite code to share with someone. The code expires in 24 hours and can only be used ... Write crow_glasses_add_to_note Append a line to an active glasses note session. If session_id is omitted, the most-recent active session f... Write crow_glasses_capture_and_attach_photo During an active note session, capture a photo via the paired glasses and attach it inline to the backing n... Write crow_identity_attest Create a signed attestation linking a per-app handle (e.g., @alice@m.example on Mastodon) to this Crow root... Write crow_iptv_add_playlist Add an M3U playlist by URL. Fetches, parses, and imports all channels. Write crow_kb_create_article Create a new article in a knowledge base collection. For multilingual pairs, create the first language vers... Write crow_kb_create_collection Create a new knowledge base collection. Collections are top-level containers that hold categories and artic... Write crow_kb_edit_article Update an existing article Write crow_kb_edit_collection Update an existing knowledge base collection Write crow_kb_import_article Import text or markdown content as a new knowledge base article. Use this to bulk-import existing guides. Write crow_kb_manage_resources Manage structured resource entries for a knowledge base article. Actions: add, edit, list, delete, flag (AI... Write crow_kb_publish_article Publish a draft article (makes it visible based on collection visibility). Uses a two-step confirmation: fi... Write crow_linkding_create Create a new bookmark in Linkding Write crow_linkding_update Update an existing Linkding bookmark Write crow_media_add_source Subscribe to an RSS/Atom feed, Google News search, or YouTube channel. Write crow_media_article_action Perform an action on an article: star, unstar, save, unsave, mark_read, mark_unread, thumbs_up, thumbs_down Write crow_media_digest_settings Configure email digest delivery settings. Write crow_media_schedule_briefing Schedule automatic briefing generation using Crow Write crow_media_smart_folders Manage smart folders: saved filter presets that auto-populate with matching articles. Write crow_miniflux_add_feed Subscribe to a new RSS/Atom feed by URL Write crow_miniflux_mark_read Mark entries as read. Pass specific entry IDs or Write crow_navidrome_playlists List or create playlists in Navidrome Write crow_notification_settings Get or update notification preferences (which types are enabled). Write crow_paperless_correspondents List all correspondents or create a new correspondent in Paperless-ngx Write crow_paperless_tags List all tags or create a new tag in Paperless-ngx Write crow_paperless_update Update document metadata (title, tags, correspondent, document type) Write crow_paperless_upload Upload a document to Paperless-ngx for OCR processing and archiving Write crow_publish_post Publish a blog post (sets status to published and records timestamp). Returns a preview and confirmation to... Write crow_react Send an emoji reaction to a Crow contact. Write crow_register_backend Register an external MCP server as a data backend. Creates a data_connector project if project_id is not pr... Write crow_register_instance Register a Crow instance in the instance registry. Each Crow installation directory is a separate instance.... Write crow_room_close Close an active companion room. Invalidates the room token and optionally notifies participants. Write crow_room_invite Invite a Crow contact to join your companion room. Generates a room token and sends an encrypted invite via... Write crow_schedule_pipeline Schedule a pipeline to run on a cron schedule. Creates an entry in Crow Write crow_send_group_message Send a message to all members of a contact group. Messages are sent as individual encrypted DMs with group ... Write crow_send_message Send an encrypted message via the Nostr network. Messages cannot be retracted once sent. Write crow_set_background Set a previously generated gallery image as the current companion background Write crow_set_discoverable Opt in or out of contact discovery by setting your email hash. Other Crow users can then find you by email ... Write crow_share_post Share a blog post with a Crow peer via P2P Write crow_shiori_save Save a URL to Shiori. Optionally archive the page content for offline reading. Write crow_shiori_update Update a Shiori bookmark Write crow_tax_add_1098 Add a 1098 form (E for student loan, main for mortgage). Write crow_tax_add_1099 Add a 1099 form (SA, INT, DIV, NEC, G, MISC). Write crow_tax_add_deduction Add a deduction (educator expense, charitable, medical, SALT, etc.). Write crow_tax_add_dependent Add a dependent to the return. Write crow_tax_add_education_credit Add a 1098-T education credit (AOTC or Lifetime Learning Credit). Write crow_tax_add_w2 Add a W-2 wage statement to the return. Write crow_tax_generate_pdfs Fill IRS PDF forms and generate a summary document. Write crow_tax_set_hsa Set HSA details (coverage type, contributions, distributions). Write crow_tax_set_self_employment Add Schedule C self-employment income and expenses. Write crow_tax_set_special Set special situations (6013(h) election, age, blindness). Write crow_trilium_clip_web Clip a web page and save it as a note in TriliumNext. Fetches the URL, extracts text content, and creates a... Write crow_trilium_create_note Create a new note in TriliumNext under a specified parent. Write crow_trilium_day_note Get or create today Write crow_trilium_update_note Update a note Write crow_update_context_section Update an existing crow.md section Write crow_update_instance Update a registered instance Write crow_update_memory Update an existing memory Write crow_update_project Update a project Write crow_update_schedule Update or delete a scheduled task by ID. Provide only the fields you want to change, or set delete=true to ... Write crow_update_setlist Update setlist metadata or reorder songs Write crow_upload_file Upload a small file (base64, <1MB) or get an HTTP upload URL for larger files. When project_id is set, the ... Write crow_vikunja_create_project Create a new Vikunja project Write crow_vikunja_create_task Create a new task in a Vikunja project Write crow_vikunja_update_task Update an existing Vikunja task (title, description, done, priority, due date) Write crow_voice_memo Send a text voice memo to a Crow contact. The recipient Write crow_wallabag_save Save a URL to Wallabag for reading later. Wallabag fetches and stores the article content. Write crow_wallabag_update Update an article: toggle starred/archived status, change tags Write forgejo_create_issue Open a new issue in a repository Write forgejo_create_repo Create a new repository owned by the token user Write fw_add_to_playlist Append tracks to an existing playlist. track_ids are integer Funkwhale track IDs (from fw_search type= Write fw_create_playlist Create a new empty playlist owned by the authenticated user. privacy_level defaults to Write fw_upload_track Upload an audio file to a library. Pass Write gitea_create_issue Open a new issue in a repository Write gitea_create_repo Create a new repository owned by the token user Write gts_import_blocklist Import a domain blocklist (IFTAS / The Bad Space / custom URL, one domain per line). QUEUED — requires oper... Write gts_post Publish a status (toot) to the fediverse. Content is public by default unless visibility is narrowed. Rate-... Write gts_report_remote Send a moderation report to a remote server about one of their accounts. Write immich_create_album Create a new photo album Write lemmy_comment Reply to a post or another comment. Pass post_id (required) and optional parent_id for replies to comments.... Write lemmy_post Create a post in a community. Either url (link post) or body (text post) must be set — or both. Rate-limite... Write maker_create_learner Create a new learner profile. Admin-only. Captures consent (parent/guardian/teacher). Stored as a research_... Write maker_save_artifact Save a learner-produced artifact (e.g. Blockly workspace XML, drawing). Guest sessions return a friendly Write maker_set_mode Switch deployment mode between solo, family, classroom. Admin-only. Downgrading family→solo refuses if more... Write maker_update_learner Update a learner Write mastodon_import_blocklist Import a domain blocklist (IFTAS / Bad Space / custom URL). QUEUED — requires operator confirmation. Rate-l... Write mastodon_post Publish a status (toot). Content is public by default unless visibility is narrowed. Rate-limited: 10/hour ... Write mastodon_post_with_media Upload an image/video and publish it as a status. Uploads via POST /api/v2/media (async processing) then PO... Write pf_import_blocklist Import a domain blocklist (IFTAS / Bad Space / custom URL). QUEUED — requires operator confirmation. Rate-l... Write pf_post_photo Upload a photo and publish it as a status. Uploads via POST /api/v1/media then POST /api/v1/statuses. Pass ... Write pt_upload_video Upload a video file (single-request; use the web UI for files >2 GB). Required: channelId (numeric) OR chan... Write wf_create_post Create a post. If collection is omitted, the post is a draft (anonymous post, not federated until published... Write wf_publish_post Move a draft post into a collection (blog), making it public and federated over ActivityPub. Rate-limited: ... Write wf_update_post Edit an existing post by ID. Rate-limited: 20/hour.
DESTRUCTIVE 53 tools
Destructive caddy_remove_site Remove a site block from the Caddyfile (matched by address) and reload Caddy. Destructive — the change is p... Destructive crow_bookstack_delete Delete a page or chapter from BookStack (irreversible) Destructive crow_campaign_delete Delete a campaign and all its posts. Requires confirmation. Destructive crow_crosspost_cancel Cancel a queued cross-post before its scheduled_at fires. Idempotent — cancelling an already-published entr... Destructive crow_delete_context_section Delete a custom crow.md section. Protected sections cannot be deleted — only disabled. Scoped overrides (de... Destructive crow_delete_file Permanently delete a file from storage. This cannot be undone. Returns a preview and confirmation token on ... Destructive crow_delete_memory Permanently delete a memory. This cannot be undone. Returns a preview and confirmation token on first call;... Destructive crow_delete_post Permanently delete a blog post. This cannot be undone — use crow_unpublish_post to revert to draft instead.... Destructive crow_delete_setlist Permanently delete a setlist. Returns a preview and confirmation token on first call; pass the token back t... Destructive crow_glasses_undo_last_append Remove the most recently appended line from an active glasses note session. Use when the user says Destructive crow_identity_revoke Sign a revocation for a previously-published attestation. The revocation is added to /.well-known/crow-iden... Destructive crow_kb_delete_article Delete an article. Uses two-step confirmation. Destructive crow_linkding_delete Delete a bookmark from Linkding. This is permanent. Destructive crow_media_playlist Manage playlists: create, list, rename, or delete. Destructive crow_media_remove_source Remove a news source subscription. Returns a preview and confirmation token on first call; pass the token b... Destructive crow_miniflux_remove_feed Unsubscribe from a feed. This removes the feed and all its entries permanently. Destructive crow_remove_backend Remove a data backend registration. This cannot be undone. Returns a preview and confirmation token on firs... Destructive crow_remove_from_setlist Remove a song from a setlist Destructive crow_reset_refcounts Operator kill switch: clear all lifecycle refcounts and reconcile against live provider health. Use when re... Destructive crow_revoke_access Revoke a previously shared item or project from a contact. Stops ongoing sync for shared projects. Returns ... Destructive crow_revoke_instance Revoke a registered instance — sets status to Destructive crow_shiori_delete Delete a bookmark from Shiori. This is permanent and removes any cached content. Destructive crow_tax_purge_return Securely delete tax return data after filing. This is irreversible. Destructive crow_trilium_delete_note Delete a note from TriliumNext. Requires explicit confirmation. Destructive crow_vikunja_delete_task Delete a task from Vikunja (irreversible) Destructive crow_wallabag_delete Permanently delete a saved article from Wallabag Destructive crowdsec_delete_decision Lift a specific ban by its decision ID. Destructive — the target IP/range will no longer be blocked by any ... Destructive fw_defederate Full defederation: block domain + purge cached content. QUEUED — requires operator confirmation in the Nest... Destructive fw_delete_playlist Permanently delete an entire playlist. Requires confirm:true and is irreversible — get explicit user confir... Destructive fw_remove_from_playlist Remove a single track from a playlist by its 0-based position (use fw_get_playlist_tracks to find the right... Destructive fw_unfollow Remove a library follow or channel subscription. Rate-limited: 30/hour. Destructive lemmy_defederate Defederate from a remote instance (block + purge cached content). Admin-only; QUEUED. Destructive lemmy_media_prune Trigger a pict-rs prune of remote media. Exposes the admin purge endpoint; deletes cached media for federat... Destructive maker_delete_learner Permanently delete a learner and cascade to sessions, transcripts, memories, and storage references. Tier-1... Destructive maker_force_end_session Hard kill a session. Skips the 5s flush; any in-flight artifact save may be lost. Requires a reason (logged). Destructive maker_revoke_batch Revoke every session in a batch (use when a printed QR sheet is lost). Admin-only. Requires a reason (logged). Destructive pf_defederate Defederate from a remote domain (block + purge cached content + sever follows). QUEUED — requires operator ... Destructive pt_defederate Full defederation: block + unfollow + purge cached videos from a remote instance. QUEUED — requires operato... Destructive wf_delete_post Permanently delete a post. Destructive — the post is gone from the local instance; federated copies on remo... Destructive gts_media_prune Manually trigger pruning of remote media older than N days. The scheduled cron (scripts/media-prune.sh) run... Destructive mastodon_media_prune Manually trigger pruning of cached remote media older than N days. The sidekiq scheduler handles this on a ... Destructive pf_media_prune Manually trigger a prune of remote media older than N days. The scheduled horizon job handles this on a rec... Destructive pt_media_prune Trigger pruning of remote-cached video files older than N days. PeerTube runs this on a scheduled job; this... Destructive crow_data_write Execute a write SQL statement (INSERT, CREATE TABLE, UPDATE, DELETE) on a user-owned database. Separate fro... Destructive crow_deactivate_server Deactivate a server Destructive crow_dismiss_all_notifications Dismiss all notifications, optionally filtered by type or date. Destructive gts_block_domain Block an entire remote domain (no federation, no media fetch). QUEUED — requires operator confirmation in t... Destructive gts_defederate Defederate from a remote domain (stop all ActivityPub interaction). Stronger than block_domain — existing f... Destructive mastodon_defederate Instance-wide defederation — admin-only. Uses the admin/domain_blocks endpoint with severity=suspend. QUEUE... Destructive matrix_leave_room Leave a room. Destructive — your messages stay but you lose access. Rate-limited: 30/hour. Destructive pf_block_domain Block an entire remote domain (no federation, no media fetch). QUEUED — requires operator confirmation in t... Destructive pt_block_server Block an entire remote instance (admin, instance-scope blocklist — hides all accounts + videos from that do... Destructive crow_frigate_set_detect Enable/disable motion detection for a camera. DESTRUCTIVE: disabling detect also stops motion-triggered rec...
EXECUTE 54 tools
Execute changedetection_recheck Trigger an immediate recheck of one watch. Returns once the request is queued (the fetch may still be runni... Execute crow_browser_capture_har Start or stop HAR (HTTP Archive) recording to capture all network requests. Useful for API discovery. Execute crow_browser_evaluate Execute JavaScript in the page context and return the result. Execute crow_browser_launch Start/restart the browser and connect via CDP. Returns the VNC URL. Pass proxy_url to route through a proxy... Execute crow_browser_navigate Navigate to a URL. Waits for page load. Injects stealth scripts. Execute crow_browser_wait_for Wait for an element to reach a state (appear, hide, attach, detach). Robust alternative to fixed sleeps for... Execute crow_browser_wait_for_user Pause automation and display a message. Waits until the user signals to continue (call again with resume=tr... Execute crow_media_refresh Trigger an immediate feed refresh for one or all sources Execute crow_orchestrate Start a multi-agent team on a complex goal (research, multi-step analysis, code work). Runs in the BACKGROU... Execute crow_plex_play Start playback of an item on a Plex client. Lists available clients if no client_id is provided. Execute crow_run_pipeline Execute a named pipeline immediately. Pipelines are predefined multi-agent workflows (e.g. memory-consolida... Execute crow_tax_calculate Run the full tax calculation and return a summary with audit trail. Execute fw_media_prune Manually trigger a prune of cached remote audio files older than N days. Default retention is 14 days (7 da... Execute fw_stop_playback Stop the currently-playing audio stream on the paired glasses and clear any queued tracks. This is a hard s... Execute maker_start_guest_session Mint an ephemeral guest session (no learner profile, no memories, no transcripts, no artifact save). 30-min... Execute maker_start_session Mint a new kiosk session for a learner and return a redemption code (NOT the raw token). The QR/URL carries... Execute maker_start_sessions_bulk Mint sessions for multiple learners sharing a batch_id. Returns an array of redemption codes for a printabl... Execute adguard_toggle_protection Enable or disable AdGuard Home protection. Disabling stops ALL DNS filtering until re-enabled — clients wil... Execute caddy_reload Validate and apply the current Caddyfile via Caddy Execute crow_browser_block_requests Block requests whose URL contains any of the given substrings (ads/trackers/images) via raw CDP Fetch. Spee... Execute crow_browser_capture_responses Capture network responses (incl. JSON bodies) via raw CDP — the fastest way to find a site Execute crow_browser_download Trigger a file download by clicking an element, and save it to the host at ~/.crow/browser-downloads/. Uses... Execute crow_browser_extract_article Extract a (possibly paywalled) article by trying a fallback ladder: live page → archive.today → Wayback → 1... Execute crow_glasses_capture_photo Ask paired glasses to capture a still photo. Returns a hint string — the photo itself arrives asynchronousl... Execute crow_glasses_confirm_continuous_recording Confirm the user Execute crow_glasses_end_note_session End a glasses note session. Runs summarization + action-item extraction via the configured AI provider, pre... Execute crow_jellyfin_control Remote control a Jellyfin playback session: play/pause, stop, seek, skip, volume Execute crow_kodi_control Transport controls: play/pause, stop, seek, skip, volume. Use command + optional value. Execute crow_kodi_play Play a library item by ID, or a direct file/URL. Specify media_type + id for library items, or file for dir... Execute crow_plex_control Transport controls for Plex playback: play, pause, stop, skip, seek, volume Execute crow_regenerate_embeddings Re-embed memories to fix stale embeddings (e.g. after content updates or after enabling semantic search). O... Execute crow_tools Route to external integration tools (Trello, Canvas, Slack, etc.) and remote Crow instances. Use crow_disco... Execute fw_block_domain Instance-wide block of a remote domain (all actors/libraries from that domain become unreachable). QUEUED —... Execute fw_block_user Block a single user (by full actor handle @user@server). Inline; rate-limited: 5/hour. Execute fw_next_track Skip to the next track in the album queue on the glasses. If no tracks remain, playback stops. Execute fw_pause Pause the currently-playing audio stream on the glasses. Playback can be resumed with fw_resume. Execute fw_play Play a Funkwhale track on the paired glasses / phone speaker. Takes a track UUID from fw_search. Returns a ... Execute fw_play_album Play every track of an album sequentially through the paired glasses speaker. Takes the album Execute lemmy_block_instance Block an entire remote instance (no federation, no fetched content). Admin-only; QUEUED — requires operator... Execute matrix_join_room Join a room by ID or alias. If the room lives on another server, Dendrite federates the join (may take seve... Execute matrix_send_message Send a text (or notice, or HTML-formatted) message to a room. Rate-limited: 20/hour. Execute crow_activate_server Activate a server Execute crow_browser_add_cookies Inject cookies into the current context (e.g. a copied session). Each cookie needs name, value, and either ... Execute crow_browser_click Click an element with position randomization for stealth. Execute crow_browser_fill_form Fill form fields with human-like typing. Accepts a map of selector -> value pairs. Execute crow_browser_scroll_extract Auto-scroll an infinite-scroll / lazy-load page, collecting items each pass. Stops when no new items load o... Execute crow_browser_set_headers Set extra HTTP headers (e.g. User-Agent, Referer) for all subsequent requests in this context. Apply BEFORE... Execute crow_campaign_publish_post Immediately publish a single post to Reddit. Requires confirmation. Execute crow_gis_create_geojson Generate GeoJSON from a SQL query result that includes lat/lng columns. Returns a FeatureCollection for use... Execute crow_glasses_speak Send a text line to be spoken through paired glasses. Requires the user to have at least one glasses device... Execute crow_wm Screen control. ALWAYS call this tool when the user wants to: open/play/watch/search something, close a win... Execute fw_resume Resume paused audio playback on the glasses. Execute maker_redeem_code INTERNAL: redeem a one-shot code for a session token. The /kiosk/r/:code HTTP handler calls this server-sid... Execute maker_unlock_idle Clear an idle-locked session without ending it. Admin-only.
FINANCIAL 3 tools
Financial crow_tax_prepare_from_documents One-step tax return preparation: creates a return from ALL confirmed documents uploaded through the Tax Fil... Financial crow_actual_create_transaction Create a new transaction. Amount in cents (negative = expense, positive = income). Financial crow_tax_set_capital_gains Add Schedule D capital gain/loss transactions.

Route Crow through PolicyLayer and every one of its 576 tools is checked against your policy before it runs.

CHECK YOUR STACK →

See every tool, the dangerous ones, and the token cost across your stack.

// FAQ
How many tools does the Crow MCP server have? +

The Crow MCP server exposes 576 tools across 5 categories: Read, Write, Destructive, Execute, Financial.

How do I enforce policies on Crow tools? +

Route the Crow server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard; they are enforced on every call before it reaches the server.

What risk categories do Crow tools fall into? +

Crow tools are categorised as Read (279), Write (187), Destructive (53), Execute (54), Financial (3). Each category has a recommended default policy.

Enforce policy on every Crow tool call.

Start from Crow, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →

Free to start. No card required.

43,000+ MCP servers and 220,000+ tools scanned and risk-classified.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.