Instagram

List the Instagram accounts connected to this install — id, username, label.

Like one or more Instagram posts, addressed by post URL or shortcode (NOT the numeric media id the official API returns). Returns a job_id; poll instagram_actions_get_job_status.

List the comments on a post, addressed by URL or shortcode. Returns each comment_id in the form that instagram_actions_like_comment accepts. The official API returns a DIFFERENT id that is not accepted there. Returns a job_id; poll instagram_actions_get_job_status.

Like one or more comments. Each comment_id MUST come from instagram_actions_list_comments. The id returned by the official instagram_get_ig_media_comments belongs to a different namespace and is rejected. Returns a job_id; poll instagram_actions_get_job_status.

Follow one or more Instagram users by @username. Returns a job_id.

Unfollow one or more Instagram users by @username. Returns a job_id.

List a specific user's currently available stories (read-only, does not mark as seen). Returns a job_id (or the result inline when wait=true).

Watch (mark as seen) the stories of one or more profiles. Omit `targets` to watch the home tray. Returns a job_id.

Send the FIRST direct message to someone who has never messaged this account (the official API cannot do this). One recipient per call, low daily cap. Returns a job_id.

Get the status/progress/result of a previously started action by its job_id. Bulk support: accepts job_ids for batched execution.

List recent action jobs for the connected account, optionally filtered by status.

Get Instagram Business Account info including profile details and statistics. Use "me" to query your own authenticated account, or provide the numeric Business Account ID of an account you manage. NOTE: followers_count and follows_count are returned for your own profile (ig_user_id="me"). For other accounts, follower/… Bulk support: accepts ig_user_ids for batched execution.

List account media. REST response: read result.result.data and result.result.paging (not result.data). Null engagement fields mean unavailable, not zero; use media insights for saved, shares and views. Bulk support: accepts ig_user_ids, auto_resolve_fb_page_ids for batched execution.

Get a published Instagram Media object (photo, video, story, reel, or carousel). Use when you need to retrieve detailed information about a specific Instagram post including engagement metrics, caption, media URLs, and metadata. NOTE: This action is for published media only. For unpublished container IDs (from INSTAGR… Bulk support: accepts ig_media_ids for batched execution.

Tool to retrieve comments on an Instagram media object. Use when you need to fetch comments from a specific Instagram post, photo, video, or carousel owned by the connected Business/Creator account. Supports cursor-based pagination for navigating through large comment lists. An empty data array in the response indicat… Bulk support: accepts ig_media_ids for batched execution.

Read media insights. REST response: read result.result.data. Request metric=[views,reach,saved,shares] for views, reach, saves and shares; availability depends on the media type and metric. Bulk support: accepts ig_media_ids for batched execution.

Read account insights at result.result.data in REST. An empty data array means insights are unavailable, not zero. It does not by itself establish a follower-count restriction. Bulk support: accepts ig_user_ids for batched execution.

Tool to create a media container for Instagram posts. Use this to create a container for images, videos, Reels, or carousels. This is the first step in Instagram's two-step publishing process - after creating the container, use the media_publish endpoint to publish it. Bulk support: accepts ig_user_ids, location_ids for batched execution.

Create a draft carousel post with multiple images/videos before publishing. Instagram requires carousels to have between 2 and 10 media items. Container creation_ids expire in under 24 hours, so publish promptly after creation. Bulk support: accepts ig_user_ids, location_ids for batched execution.

Tool to publish a media container to an Instagram Business account. This action automatically waits for the container to finish processing before publishing. Instagram enforces an account-specific content-publishing rate limit over a 24-hour moving window; the exact cap varies by account and should be read live via IN… Bulk support: accepts ig_user_ids, creation_ids for batched execution.

Tool to create a reply to an Instagram comment. Use when you need to reply to a specific comment on an Instagram post owned by a Business or Creator account. The reply must be 300 characters or less, contain at most 4 hashtags and 1 URL, and cannot consist entirely of capital letters. Bulk support: accepts ig_comment_ids for batched execution.

Send a text message to an Instagram user via DM in an existing conversation. Cannot initiate new DM threads — a prior conversation must exist. Requires an Instagram Business or Creator account with messaging permissions. Fails with error_subcode 2534022 if outside the messaging window; do not retry these failures. Bulk support: accepts ig_user_ids, recipient_ids, reply_to_message_ids for batched execution.

Send an image via Instagram DM to a specific user. Each send modifies inbox state; avoid bulk or automated sends without explicit user approval. Bulk support: accepts ig_user_ids, recipient_ids for batched execution.

Mark Instagram DM messages as read/seen for a specific user. Sends a 'mark_seen' sender action to indicate messages from the specified recipient have been read. Marking as seen is visible to the other party and changes inbox read state — use with explicit user approval in automated or bulk flows. IMPORTANT LIMITATIONS… Bulk support: accepts ig_user_ids, recipient_ids for batched execution.

Get details about a specific Instagram DM conversation (participants, etc). Requires a Business or Creator account with Instagram messaging permissions; personal accounts will return permission errors. Newly sent/received messages may take a few seconds to appear in results. Bulk support: accepts conversation_ids for batched execution.

List all Instagram DM conversations for the authenticated user. Requires a Business/Creator account with messaging permissions; personal accounts return empty results. Response conversations are nested under `data.data` — accessing top-level `data` as the final list returns zero items. An empty `data` list is a valid … Bulk support: accepts user_ids, ig_user_ids for batched execution.

List all messages from a specific Instagram DM conversation. Requires a Business or Creator account with messaging permissions; personal accounts return empty results. Response data is nested under data.data (double-wrapped); attachment-only messages may have empty text fields. Bulk support: accepts conversation_ids for batched execution.

Show the current MCP platform and adapter versions.

无需输入参数。

Report a bug, missing feature, or send feedback. Include the conversation array with recent messages for reproduction.

Returns connection status and URLs. When all providers are connected, returns authenticated:true and empty pending[]. When credentials are missing, returns connect_url for the toolkit and per-install URLs.

无需输入参数。

Returns the current toolkit state: installed MCPs, their connection status, the accounts connected to each one, and how many catalog tools each exposes.

无需输入参数。

The official mcp.ai marketplace — the in-platform catalog of every MCP/tool, AND the way to run them. Covers capability requests like "find an MCP that does X", "consulta um CPF", "is there a tool for Y". Core flow: action=search discovers MCPs by intent → describe returns one MCP's full profile (every tool with its id + params, pricing, auth) so you pick the right tool_id → invoke RUNS that tool. KEY: invoke works even when the MCP is NOT installed — it runs the tool pontualmente (one-off), without adding the MCP to the toolkit and without bloating the tool list. If the MCP needs a credential/login, invoke returns a connect link; if it is paid and the wallet is empty, invoke returns a checkout/top-up link (the user opens it, then you retry). Use install only to make an MCP PERMANENT in the active toolkit (its tools then show up natively in future sessions); prefer invoke for a single/occasional use. list_tools lists what is callable right now. subscribe/cancel handle per-MCP billing; report_bug sends feedback; request_mcp asks us to build a NEW MCP when nothing fits. Search/describe flag installed_in_toolkit vs installed_in_workspace. Writes (install/uninstall/subscribe/cancel and the one-off install behind invoke) require workspace owner/admin. It also carries the mcp.ai PROMPT LIBRARY, which is about ready-made prompt TEXT rather than MCPs: search_prompts finds one, get_prompt returns its full text with {{variables}} filled, and publish_prompt saves a prompt and returns a shareable mcp.ai/p/<slug> link that opens without login.

MCP.AI for IDE agents (Cursor, etc.): log in in the browser, copy the access token. Best: add it to this server's config as a header `Authorization: Bearer <token>` for a permanent, non-expiring connection. Or paste it here for a session-only login: call with { token: "<jwt>" } after the user pastes, or with no args to get the link.