Arcmira: YouTube Transcript Search

Arcmira MCP for YouTube and podcast transcript search

Search YouTube video and podcast transcripts, retrieve timestamped quotes, find entity mentions and media appearances, inspect organic endorsements and ad reads, and monitor new mentions with your agent.

Arcmira · API docs · MCP setup

Connect Arcmira

Add the remote MCP server at https://mcp.arcmira.com/mcp and sign in through your host. For setup instructions, use Connect your agent.

For a coding agent, paste:

Fetch and execute the appropriate instructions to set me up for Arcmira from https://arcmira.com/agent-setup/prompt.md

Run your first search

Ask your agent:

Search YouTube and podcast transcripts for GPU financing over the last 60 days. Return the relevant quotes with speakers when available, publication dates, transcript sources, and timestamped links. Read the surrounding exchanges before summarizing what the sources imply.

The agent calls arcmira_describe once to load the client reference, then runs a program with arcmira_execute_read. Return structured fields for further analysis and a compact, sourced answer for the reader.

const query = "GPU financing";const hits = await arcmira.search({  query, after: arcmira.daysAgo(60), before: arcmira.today(), limit: 5,});return { query, window: hits.window, passages: hits.chunks };

The search supports up to 20 passages, but full passages can exceed the execute tool's output budget. If the result is truncated, request fewer passages or return only the fields needed for the next step. Keep selected text intact and retain its source link.

Choose the method for the task

Resolve names before applying entity or channel ID filters. Plain keyword search does not need a resolver call. The MCP client's argument names differ from the HTTP and SDK names; use the reference returned by arcmira_describe.

Continue the research

Continue the investment research workflow. You will learn how to follow a search result into competing explanations, inspect the surrounding exchange, verify claims against disclosures, and write a brief with the questions that remain unresolved.

Speaker identification and commercial classifications can be incomplete. Check the linked source. Missing a source or a capability your workflow needs? Use arcmira_feedback to describe it, why it matters, and what better support would be worth. Help shape the index.

Plugin and skills · Share plugin feedback · Update installed skills

Authentication options

Two ways in. Hosts that speak the MCP authorization spec sign you in; everything else sends a key.

Sign in through the host. Add https://mcp.arcmira.com/mcp with no key. The server answers 401 with an OAuth challenge, the host registers itself against api.arcmira.com, opens arcmira.com for sign-in and consent, and connects with a token that carries the permissions you allowed. Tokens refresh on their own; revoke a host under Settings, Connected apps.

claude mcp add --transport http --scope user arcmira https://mcp.arcmira.com/mcp

codex mcp add arcmira --url https://mcp.arcmira.com/mcp

Claude Desktop, claude.ai, ChatGPT, and Cursor: add the URL as a custom connector or MCP server with no headers and follow the sign-in prompt. Per-host steps are on https://arcmira.com/agent-setup.

Send a key. Any client that cannot do the sign-in sends a bearer token instead, and the server skips OAuth.

An account key (arc_sk_...) comes from https://arcmira.com. Plan and scopes decide what each tool returns. With no account and no browser, sign up from the API. Post an email address, then post the six digit code from that inbox back.

curl -X POST "https://api.arcmira.com/v1/signups?src=mcp-tool" \  -H 'Content-Type: application/json' \  -d '{"email":"[email protected]"}'
curl -X POST "https://api.arcmira.com/v1/signups/verify" \  -H 'Content-Type: application/json' \  -d '{"email":"[email protected]","code":"482913"}'

claude mcp add --transport http --scope user arcmira https://mcp.arcmira.com/mcp --header "Authorization: Bearer $ARCMIRA_API_KEY"

{  "mcpServers": {    "arcmira": {      "url": "https://mcp.arcmira.com/mcp",      "headers": { "Authorization": "Bearer arc_sk_..." }    }  }}

The 401 body carries that signup call under error.data.unlock.action, so an agent that cannot sign in can create an account and reconnect. Discovery: https://mcp.arcmira.com/.well-known/oauth-protected-resource names the authorization server; https://api.arcmira.com/.well-known/oauth-authorization-server lists its endpoints.

Tools

Every tool also takes an optional intent, at most 300 characters: the user's request in a few words. A host that omits it loses nothing. See What we log.

Only arcmira_describe is annotated read-only. Both program tools can start Premium transcription, which uses credits from the plan and then the on-demand budget, so neither is read-only. Only arcmira_execute_write is destructive, because it can overwrite monitor settings and pause monitors; a Premium read only adds. Both can request arbitrary public videos and carry open-world annotations. Feedback adds a stored record and is an additive write. These labels describe possible effects; the execute_read name identifies the client access level, not a guarantee of no side effects.

Metered reads use credits from your plan, then the on-demand budget you set in the dashboard. Arcmira enforces that budget on every read, and no tool takes a spending amount. A search uses 4 credits per passage it returns past the first 5, a captions read 4 credits per started 15 minutes of its window, and a Premium transcript 300 credits per started 15 minutes of the whole video. A Premium read that starts a transcription reports its charge (credits, and whether they came from the plan, on-demand usage, or both). arcmira.quote(video) returns that price beforehand and uses no credits; it is a price check, not a spoken quote. When the plan or budget blocks a read, the tool result is an error (isError: true) with the API's code (paid_plan_required, quota_exceeded, spend_limit_exceeded, member_limit, fair_use_cap or on_demand_paused), a message that states what was reached and what lifts it, and unlock when a plan or setting change lifts it. The server gives the agent these facts, not a script for what to say.

A Premium transcript is one read in arcmira_execute_read:

const t = await arcmira.transcript("cdLeJU_1UH8", { quality: "premium" });return t.state === "ready" ? t.lines : { state: t.state, eta_seconds: t.eta_seconds, note: t.note };

When the video is not transcribed yet, GET /v1/transcripts/{video_id}?quality=premium starts transcribing it and answers 202 with the job and Retry-After. The client reads again at Retry-After until 22 seconds into the program, a wait all Premium reads in one program share. Reading again never uses credits twice. Still pending after that: the result carries eta_seconds, and the same read in a later program returns the lines. A failed transcription comes back as state failed or refunded with the reason, and the read does not start another unless it passes retry: true, which an agent sends only when the user asks. A plan without Premium throws paid_plan_required; a budget that blocks the read throws spend_limit_exceeded or quota_exceeded, and the price rides on the error as quote. start and end never lower a Premium charge.

Results that are empty, an error, truncated or a resolve ask carry call_id (mcpc_...), which arcmira_feedback takes to name the call.

The client covers the same tasks as the CLI and SDKs. Argument names differ; use these MCP signatures:

arcmira.today() and arcmira.daysAgo(n) give ISO dates from the server clock for date windows. Windows are half-open, as on the API: after is the first day counted and before the first day left out, so August is { after: "2026-08-01", before: "2026-09-01" }. Every dated result echoes window: { after, before }. Option names are the client's; request and response fields are the API's, so a monitor takes notify_frequency because monitors.list() returns it.

Filters take verbatim ids only. A name where an id belongs throws id_required before any network call, and the message names the fix. Resolve first, then query. resolve answers best (the name means one row), suggested (one row stands out; say you assumed it and why), or ask (several fit; let the user pick):

const r = await arcmira.resolve("Linear");const e = r.best ?? r.suggested;if (!e) return { ask: r.ask };const m = await arcmira.momentum(e.id);return { entity: e.name, id: e.id, assumed: Boolean(r.suggested), why: r.suggested?.evidence ?? null, verdict: m.verdict, last30: m.volume.mentions_30d, prior30: m.volume.mentions_prior_30d, as_of: m.as_of };

Example prompts once the server is connected:

• Which brands sponsor both TBPN and the All-In Podcast?
• How hot is Linear, the project management tool, on the spoken web right now?
• Find a moment on TBPN from the last 90 days where someone talks about stablecoins, and quote it.
• Who is speaking in the first minute of the latest TBPN episode, according to the Premium transcript?
• On TBPN, how many episodes mentioned Cursor in July versus August?

Good first ids: TBPN is channel UC-DRzaGnL_vtBUpCFH5M0tg, All-In Podcast is UCESLZhusAkFfsNsApnjF_Cg, Ramp is ent_14.

The sandbox

arcmira_execute_read and arcmira_execute_write run the program in a fresh Dynamic Worker isolate. The isolate's only network is the parent's outbound proxy, which adds the caller's credential to what it forwards, so the program never holds the key, and refuses anything outside the tool's allowlist with outbound_refused:

• arcmira_execute_read: GET https://api.arcmira.com/v1/*. A Premium transcription starts inside the transcript GET.
• arcmira_execute_write: the read set, plus POST and PATCH under /v1/monitors and PATCH under /v1/trackers. Never DELETE, and never the webhook secret rotation. A tracker is created inside a monitor, so nothing posts to /v1/trackers.

The proxy enforces this in the Worker, so a raw fetch() gets the same answer as a client method. A write method called from arcmira_execute_read throws write_tool_required before any request. The isolate gets 5 seconds of CPU, the tool waits 30 seconds of wall time, the client stops at 40 API calls with call_budget, and the rendered output is cut at 12,000 characters in total, 3,000 per string and 100 items per array, with truncated_arrays naming each cut array and a recovery line that says how to get every row. A syntax error comes back as syntax_error with the function-body rule; a thrown error as program_error with its message.

The result is valid bounded JSON with ok and value or error first, then actual calls, rate_limit, api_build, truncation facts and capped logs. Large results retain continuation and recovery fields. Execution metadata uses the same meter. Timeout reports calls: null and outcome_uncertain: true; in-flight reads may still finish and consume rows. Authentication 429/503 asks clients to retry with the same credential; only invalid credentials trigger reconnect.

Gates

A gate inside a program throws an ArcmiraError with the API's error fields, and the execute tools return it as ERROR with isError: true:

{"ok":false,"error":{"code":"recommendations_not_enabled","message":"Sponsor recommendations require Pro.","gate":"plan"},"calls":1,"outcome_uncertain":false}

Switch on code. unlock.url names the plan or setting that lifts a gate, and rate_limited carries retry_after_seconds. A 200 that withheld something (Premium transcript text, the paid-versus-organic split, sponsors past the free slice) is a normal result carrying the same body under access. The full catalog is at https://arcmira.com/docs/errors.

Every result, gates included, carries the key's budget after the call under _meta["arcmira.com/rate_limit"] as { "limit": 20, "remaining": 17, "reset": 1788819360 }, read from the API's RateLimit headers, and _meta["arcmira.com/build"]: server, deploy, api (the API build that answered) and client (the host's name from its handshake, forwarded to the API as x-arcmira-client).

What we log

Each tool call is logged to Arcmira's product analytics (PostHog), linked to the account that made it, so we can see how the tools are used and fix what fails. One record per call holds:

• the tool name, the host's name and version, the server version, and how long the call took;
• the input: the program text of either execute tool (first 4,000 characters), the arcmira_describe topic, or the arcmira_feedback category and note;
• intent, when the agent sends it;
• the outcome, not the result: ok or the error code, the result size, whether it was truncated, and the API routes the call made (GET /v1/search, not its query string).

Before the record is stored, the API replaces anything that looks like a credential or an email address (arc_ keys, Bearer tokens, sk- keys, addresses) with [redacted]. Results, transcript text and logs printed by a program are never logged. A call with no valid credential is not logged. The record is sent after the result, so logging never delays or changes an answer; a record that fails to send is dropped.

Upgrading

Upgrade notes for every release, the 0.7 tool renames and the 0.10 field changes included, are in CHANGELOG.md.

Plugin

plugins/arcmira bundles the MCP URL and six skills for Claude Code, Codex, Cursor, Agent Plugins hosts and Gemini CLI:

Every skill starts from names, never ids: each program resolves the name, returns ask options when several entities fit and none stands out, says when it assumed one, and names the entity it used. Every skill ends with the monitor line (monitors can follow what it found, and what an alert uses) and the feedback line. The skills are generated from src/reference.ts and src/skills.ts by scripts/build-skill.ts; CI fails when a file drifts or a program calls a method the reference does not document, and pnpm examples:check runs every program against production.

claude plugin marketplace add arcmira/mcpclaude plugin install arcmira@arcmira

Stay up to date

Arcmira ships changes weekly. Keep auto-update on.

• The MCP server needs nothing. It is remote: hosts fetch its tools and instructions on connect, and arcmira_describe returns the reference from the server on every call, opening with a version line. The MCP server alone is always current.

• Claude Code plugin. Auto-update is off by default for a third-party marketplace. Turn it on: run /plugin, open Marketplaces, pick arcmira, and choose Enable auto-update. Or set it in ~/.claude/settings.json:

  json

  { "extraKnownMarketplaces": { "arcmira": { "source": { "source": "github", "repo": "arcmira/mcp" }, "autoUpdate": true } } }

  Update now: claude plugin update arcmira@arcmira.

• Codex plugin. codex plugin marketplace upgrade arcmira, then restart Codex.

• Skills installed with npx skills add arcmira/mcp. npx skills update.

Each release bumps the version in every plugin manifest (a test enforces it), because Claude Code only updates a plugin whose version changed.

Gemini CLI

Install the remote connection and core Arcmira reference from this repository:

gemini extensions install https://github.com/arcmira/mcp

The root gemini-extension.json points to the same remote MCP and loads the arcmira skill as its context file, like the plugin bundle. Sign in through the MCP authorization flow. The root extension does not separately install the five optional task skills.

Develop

pnpm installpnpm dev            # wrangler dev on :8790 with a local Worker Loaderpnpm test           # node:test, sandbox programs run under a fake loaderpnpm typecheckpnpm manifest:check # every client call matches the live OpenAPI documentpnpm skill:check    # every plugins/arcmira/skills/*/SKILL.md matches src/reference.ts and src/skills.tsARCMIRA_KEY=arc_sk_... pnpm examples:check   # every worked program and read task-skill program runs against productionpnpm sandbox:check  # src/sandbox/client-source.ts matches src/sandbox/client.jsARCMIRA_KEY=arc_sk_... node --experimental-strip-types scripts/smoke.ts http://localhost:8790/mcp

src/reference.ts is the steering surface: the arcmira_describe text, the server instructions and the arcmira skill all come from it.

Copyright Arcmira. All rights reserved for the server (see LICENSE); plugins/arcmira is Apache-2.0.