Anam AI Avatars MCP Server for ChatGPT
By Anam
Overview
Anam makes personas: real-time AI avatars that talk with people over video, for uses such as customer support, sales, reception, tutoring and interview practice. Connect your Anam account to build and manage them from ChatGPT. With Anam you can: - Create a persona from a short brief, with a recommended avatar, voice and language model, and get a link to talk to it. - Browse the avatars, voices and language models your organization can use. - Add knowledge from text or web pages so a persona can answer questions about your product. - Create and manage share links. - Find past sessions by date, persona or what was said, and read their transcripts. - Look up how to add a persona to a website or app, from Anam's documentation. Who it's for: teams that build or run Anam personas, and developers adding them to their products. Limitations: you need an Anam account. Each connection acts on one Anam organization that you choose when you sign in, and reaches only that organization. Read-only access still reads everything in it, including session transcripts and knowledge. Full access can also change and delete things and start billed work. Conversations with a persona happen in a browser or your own app, not inside ChatGPT. Only use images and voices you have the right to use, and don't recreate a real person without their consent.
How to connect
- ChatGPT: Open the Anam AI Avatars listing in the ChatGPT app directory and choose Connect. ChatGPT listing Anam AI Avatars exposes 87 tools here.
Tools
anam_app_avatars
ChatGPTanam_app_personas
ChatGPTanam_app_set_persona_media
ChatGPTprevious.restore has what was reset, for the app’s Undo to send back.anam_app_talk_link
ChatGPTanam_app_voices
ChatGPTanam_avatar_create
ChatGPTimageUrl); this server cannot read files from the user’s computer. Image moderation and creation capacity limits apply. Only use images of people who have consented. Check status with anam_avatar_get after about 10 seconds.anam_avatar_delete
ChatGPThard: true also erases its stored image and versions (for data-erasure requests; it also purges a soft-deleted one), except assets another avatar uses. A hard delete is refused with 409 and the persona IDs while personas use the avatar: move them with anam_persona_update first.anam_avatar_get
ChatGPTstatus (processing until its preview videos exist, then ready), name, description, tags, style, image and versions. Its id is a persona’s avatarId.anam_avatar_list
ChatGPTsearch, tags and renderStyle; no match is an empty data list. An avatar’s id is a persona’s avatarId; anam_options lists avatars, voices and LLMs with recommendations.anam_avatar_update
ChatGPTdisplayName is editable.anam_avatar_video_cancel
ChatGPTanam_avatar_video_create
ChatGPTpersonaId; 2. avatarId, voiceId.anam_avatar_video_delete
ChatGPTanam_avatar_video_get
ChatGPTstatus and, once completed, content.url, a link to the video that works for an hour (content.expiresAt); read the job again for a new one. A failed job’s failure says why and whether a retry can fix it (retryable).anam_avatar_video_list
ChatGPTstatus (pending, running or finalizing while rendering; completed, failed or cancelled once ended), length, failure and times. Filter by status and creation dates; pass nextCursor as cursor for the next page. anam_avatar_video_get returns a finished video’s link.anam_avatar_video_retry
ChatGPTretryOf names the original, and returns that job. A job still rendering, or one whose failure.retryable is false, can’t be retried. Poll anam_avatar_video_get with the new id.anam_connection_info
ChatGPTanam_docs_read
ChatGPTcat /javascript-sdk/quickstart.mdx (use the path anam_docs_search links to, ending in .mdx). tree / -L 2 lists the pages; head -120 <page> reads the start; rg -il "livekit" / finds pages that mention a word. A read-only shell over the docs pages and OpenAPI spec. Sent to Mintlify, which hosts Anam’s docs: send documentation terms, not customer content or secrets.anam_docs_search
ChatGPTanam_integration_guide
ChatGPTanam_knowledge_document_create_from_text
ChatGPTanam_knowledge_document_create_from_url
ChatGPTanam_knowledge_document_delete
ChatGPTanam_knowledge_document_download
ChatGPTdownloadUrl, a link to the document’s original file that works for an hour, and its filename.anam_knowledge_document_get
ChatGPTknowledgeFolderId), chunking and status. Only READY documents are searchable; a FAILED one has errorMessage.anam_knowledge_document_list
ChatGPTstatus. Only READY documents are searchable; a FAILED one has errorMessage.anam_knowledge_document_update
ChatGPTfilename is editable; the stored file and its extracted chunks are unchanged. Chunk settings can only be set at upload, so a non-null chunkSize or chunkOverlap returns 400.anam_knowledge_folder_create
ChatGPTid. Add documents with anam_knowledge_document_create_from_text or anam_knowledge_document_create_from_url, then let a persona search it with anam_persona_knowledge_add.anam_knowledge_folder_delete
ChatGPTanam_knowledge_folder_get
ChatGPTanam_knowledge_folder_list
ChatGPTanam_knowledge_folder_search
ChatGPTREADY documents by meaning and returns the closest chunks, at most limit. Up to 60 searches a minute per organization, across keys and folders.anam_knowledge_folder_update
ChatGPTanam_llm_delete
ChatGPTanam_llm_get
ChatGPTurls. API keys are never returned.anam_llm_list
ChatGPTid is a persona’s llmId.anam_meeting_invite_cancel
ChatGPTcancelled, ended or failed succeeds too.anam_meeting_invite_create
ChatGPTpersonaId, to join a Google Meet, Zoom or Microsoft Teams meeting. The invite is created in pending status. displayName gets an AI disclosure suffix before the persona joins, and the persona’s video shows an AI label. joinAt schedules the join; at least 10 minutes ahead reserves the persona a place.anam_meeting_invite_get
ChatGPTstatus (pending, active, ended, cancelled or failed, with statusReason), joinState, join time and, once the persona joins, sessionId.anam_meeting_invite_list
ChatGPTstatus, such as pending (not joined yet) or active (in a call).anam_options
ChatGPTsearch matches names, descriptions and tags: use broad words (woman, office, friendly); roles such as receptionist rarely match. Voices list their languages. Use the IDs with anam_persona_create, anam_persona_update or anam_quickstart.anam_persona_create
ChatGPTid; anam_share_link_create gives it a link anyone can open to talk to it. systemPrompt is saved into brain. zeroDataRetention: true needs an enterprise plan and a ZDR-compatible voice, LLM and voiceDetectionOptions.sttProvider. anam_quickstart (your prompt or a role) and anam_persona_from_brief (a description) instead pick defaults and add a share link.anam_persona_delete
ChatGPTKnowledge_<persona> tool from anam_persona_knowledge_add is left unused: delete it with anam_tool_delete unless another persona uses it.anam_persona_from_brief
ChatGPTanam_persona_get
ChatGPTbrain), avatar, voice, LLM, toolIds and tools, knowledge, share links and session settings.anam_persona_knowledge_add
ChatGPTanam_persona_knowledge_remove
ChatGPTanam_persona_list
ChatGPTmeta: a summary of each (name, avatar, voice, llmId, language, description to 200 characters); anam_persona_get has the rest in full. search matches names.anam_persona_update
ChatGPTavatarId may reset avatarModel if the avatar doesn't support it, and a new voiceId resets voiceGenerationOptions unless new ones are sent. voiceGenerationOptions sent with the current voiceId are merged into the saved ones (so a key can't be removed that way), sent without voiceId replace them, and null clears them. They must suit the voice's model, or the model override when set. Published at once, not as a draft: new sessions and share links use it straight away. toolIds replaces the whole tool list, knowledge tools included: send every ID to keep. anam_persona_knowledge_add and anam_persona_knowledge_remove attach and detach knowledge folders.anam_persona_widget_get
ChatGPTanam_pronunciation_dictionary_create
ChatGPTvoiceGenerationOptions.pronunciationDictionaryId. Entries are validated as a whole: if any entry is invalid, nothing is saved and the response lists every error by one-based row.anam_pronunciation_dictionary_delete
ChatGPTDICTIONARY_ATTACHED and the persona IDs; remove it from those personas in Lab first.anam_pronunciation_dictionary_get
ChatGPTanam_pronunciation_dictionary_import
ChatGPT{ text, pronunciation, case_sensitive? } rows, with an optional Anam sounds_like fallback. A pronunciation containing << or >> is imported as an IPA entry, anything else as a sounds-like one. The file must hold 1 to 500 rows in at most 1 MiB of UTF-8 JSON (an array is measured without whitespace). If any row is invalid, nothing is saved and every error is listed by row.anam_pronunciation_dictionary_list
ChatGPTanam_pronunciation_dictionary_update
ChatGPTname, entries or both and returns it; entries replaces the whole list. The save is live: sessions that start afterwards use the new entries, and running ones keep theirs. With expectedRevision, the update is refused with 409 REVISION_CONFLICT if the dictionary was saved since; without it, the last write wins. If any entry is invalid, nothing is saved and every error is listed by row.anam_quickstart
ChatGPTsystemPrompt or else the template’s prompt, and a share link (up to 10 sessions over 7 days, on the organization’s minutes). Returns the persona, the link’s URL and its Lab page.anam_scaffold
ChatGPTANAM_API_KEY environment variable for a short-lived session token, a page using @anam-ai/js-sdk, an env file without the key, and install and run commands. express and fastapi come with a Vite page.anam_session_analytics_get
ChatGPTincludeMessages: true and transcripts on.anam_session_concurrency_get
ChatGPTanam_session_delete
ChatGPTscope all (default): the report (transcript, analytics, tool calls), insights, summary, recording, initial message and requester IP; recording: the video only; report: the report, insights and summary only. The session record and usage stay for billing. Irreversible; repeating it succeeds.anam_session_get
ChatGPTtoolCalls (never for LiveKit or audio-passthrough sessions): names, timing, status and errors, with arguments and results only when transcripts were on and the session isn’t zero-data-retention. What was said is in anam_session_transcript_get.anam_session_insights
ChatGPTfindings): start here to judge one call or a period. With sessionId: reply speed, interruptions, slowest turns, failed tool calls, errors and the first and last messages. Without: a roll-up by date and persona (last 24 hours by default, up to 31 days) of durations, exit statuses, latency and failure rates. CLOSED_BY_ENGINE includes a normal end of call by the user or app.anam_session_list
ChatGPTmeta: a summary of each (persona, client label, start, end, length, exit status); anam_session_get has the rest. CLOSED_BY_ENGINE includes a normal end of call by the user or app. anam_session_search finds calls by what was said; anam_session_insights says how they went.anam_session_recording_get
ChatGPTanam_session_search
ChatGPTtext, returns a compact list of sessions, newest first. With text, also reads the transcripts of up to 25 of the newest matching sessions and returns those that mention it, with snippets; narrow the date range to search further back. With sessionId, reads that session’s whole transcript instead, 50 messages at a time (limit, up to 100): pass the returned nextOffset as offset until it is null.anam_session_stats
ChatGPTsessionCount only. It reads the 1,000 newest matching sessions with analytics; truncated says more matched. The last 24 hours by default, up to 31 days. anam_session_insights summarises the same with findings.anam_session_stop
ChatGPTendTime, exit status and report follow shortly after, so anam_session_get may briefly still show it active.anam_session_transcript_get
ChatGPTlimit (up to 100) says otherwise: pass the returned nextOffset as offset until it is null. totalMessages counts them all. Zero-data-retention sessions have no transcript.anam_share_link_create
ChatGPTurl. Anyone with the link can talk to the persona until it expires (expiresInHours, 168 by default) or runs out of uses (usageLimit, 10 by default).anam_share_link_delete
ChatGPTurl stops working.anam_share_link_get
ChatGPTurl, persona, expiry, usage, allowed origins and whether it is enabled.anam_share_link_list
ChatGPTmeta: the public url that opens the persona in a browser, persona, expiry, usage and whether it is enabled. Filter by personaId or search.anam_share_link_update
ChatGPTanam_tool_delete
ChatGPTusageCount (anam_tool_get) says how many personas use it; take it out of their toolIds first. Anam’s system tools can’t be deleted.anam_tool_get
ChatGPTusageCount. A webhook config shows its structure, not its values: header values, URL IDs and tokens, query values and body template values show as •••• (a Bearer scheme, {{placeholders}}, header names, JSON keys and the host stay). Send them back unchanged to keep them.anam_tool_list
ChatGPTmeta: name, type (SERVER_RAG is knowledge), config and usageCount (personas using it). A persona uses the tools whose IDs are in its toolIds.anam_usage_get
ChatGPTanam_voice_add
ChatGPTanam_voice_add_status
ChatGPTpending, completed with the new voiceId (and personaUpdated when the link named a persona this connection can read), cancelled or expired.anam_voice_clone
ChatGPTid is a persona’s voiceId. Only clone the voice of someone who has consented. Cloning runs on Anam’s voice provider and needs a plan with voice cloning. Each call creates another voice, so check anam_voice_list before retrying. It shares the organization’s hourly limit for voice imports and model changes.anam_voice_delete
ChatGPTanam_voice_get
ChatGPTcountry), description and sample URL. Its id is a persona’s voiceId.anam_voice_import
ChatGPTid is a persona’s voiceId. The voice is saved once a speech sample generates with it: up to 95 seconds, and about a minute for a first ElevenLabs Voice Library import. displayName and description default to the provider’s. Only use voices of people who have consented. An import creates a new voice and generates a speech sample on Anam’s provider account, or with useOrganizationKey: true on the organization’s saved ElevenLabs key and its credits (for its own clones and designed voices). Retrying creates another voice, so check anam_voice_list first. A shareUrl (instead of provider and providerVoiceId) imports a Professional Voice Clone shared by link; one the organization already imported on that model gets a new sample instead of a new voice.anam_voice_list
ChatGPTcountry), languages and sample URL. Filter by search, language, gender and country; no match is an empty data list. A voice’s id is a persona’s voiceId.anam_voice_update
ChatGPTCapabilities
Example Prompts
Click any prompt to copy it.
Links
Similar Apps
Listed on MCP App Store
Add this badge to your site or README. It links to this page.
HTML
<a href="https://mcp-app-store.keenethics-labs.com/app/anam-ai-avatars"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>
Markdown
[](https://mcp-app-store.keenethics-labs.com/app/anam-ai-avatars)





