Skip to content

Beautiful, AI-native markdown IDE and LLM wiki

Document meeting ingestion into Open Knowledge (#2540)

tim-inkeep47019dac8fe8
diff --git a/.changeset/entity-vault-meeting-ingest.md b/.changeset/entity-vault-meeting-ingest.md
new file mode 100644
index 000000000..4949d8605
--- /dev/null
+++ b/.changeset/entity-vault-meeting-ingest.md
@@ -0,0 +1,5 @@
+---
+"@inkeep/open-knowledge": patch
+---
+
+The `entity-vault` starter pack skill now covers ingesting meetings from a recorder. When a meeting-recorder MCP (Granola, Fireflies, Circleback, tl;dv, Fathom) is registered alongside `open-knowledge`, the agent pulls recent meetings and writes each one into `meetings/`, addressed as `meetings/<source>-<source_meeting_id>` so re-syncing a meeting updates it in place instead of creating a duplicate. Meeting notes ingested this way carry `source:` and `source_meeting_id:` frontmatter, which is the dedup key. Transcripts stay verbatim, and the recorder is bring-your-own, so the same behavior works on demand or driven by a scheduler. The pack's `meeting` template now carries the `source` and `source_meeting_id` keys and a `## Transcript` section, so a meeting you author by hand and one pulled from a recorder share the same shape.
diff --git a/docs/content/get-started/overview.mdx b/docs/content/get-started/overview.mdx
index 1beea7ec4..da98fe7c0 100644
--- a/docs/content/get-started/overview.mdx
+++ b/docs/content/get-started/overview.mdx
@@ -23,6 +23,7 @@ OpenKnowledge is three layers working together: a "WYSIWYG" editor for you, MCP
- A **built-in TUI** in the desktop app for Claude, Codex, OpenCode, and Cursor
- Graph and wiki link visualizations
- **Embeddable HTML** for writing engineering specs and visualized reports.
+- **Meeting ingestion** from Granola, Fireflies, and other recorders. On demand in any editor, or automatically with the [OpenClaw](../integrations/openclaw.mdx) gateway. See [Ingest meetings](../workflows/meeting-ingestion.mdx).
## Where to start
diff --git a/docs/content/integrations/openclaw.mdx b/docs/content/integrations/openclaw.mdx
index 2859b3fe6..22cc3ae40 100644
--- a/docs/content/integrations/openclaw.mdx
+++ b/docs/content/integrations/openclaw.mdx
@@ -6,6 +6,8 @@ icon: custom/OpenClaw
[OpenClaw](https://openclaw.ai) is an agent gateway: it runs agents that connect to MCP servers. Once OpenKnowledge is registered as one of them, every agent OpenClaw runs can read, search, and write your knowledge base through OK's tools — with full attribution, link-awareness, conflict-awareness, and a live browser preview. Inside an OK project, agents work markdown through those tools rather than native file access, so every edit stays attributed and previewable.
+**Auto-sync your meetings.** The gateway has a built-in scheduler and a webhook listener, so once OpenKnowledge is registered it can run tasks against your knowledge base unattended. The headline example is keeping meeting notes flowing in on their own. See [Ingest meetings](../workflows/meeting-ingestion.mdx) for the cron recipe and the recorder list. The webhook mechanics, for running an agent the moment a meeting event fires, are documented below.
+
## Install
<McpInstall editor="OpenClaw">
@@ -99,6 +101,50 @@ If the `open-knowledge` server's tools aren't in the agent's tool list, run tool
That sequence exercises every layer — discovery, config, read, write, link graph, search, and preview — in one pass.
+## Trigger runs from external events
+
+Besides the cron poll (see [Ingest meetings](../workflows/meeting-ingestion.mdx)), the gateway can run an agent the moment something happens elsewhere. Enable the hooks listener in the gateway config:
+
+```json5
+{
+ hooks: {
+ enabled: true,
+ token: "shared-secret",
+ path: "/hooks",
+ },
+}
+```
+
+The `token` is a dedicated secret, separate from your gateway auth token. Every request must carry it as `Authorization: Bearer <token>` (recommended) or `x-openclaw-token: <token>`. Query-string tokens are rejected.
+
+| Endpoint | What it does |
+| --- | --- |
+| `POST /hooks/agent` | runs an isolated agent turn |
+| `POST /hooks/wake` | enqueues a system event into the main session |
+| `POST /hooks/<name>` | resolves through `hooks.mappings`, to reshape a vendor payload into a prompt |
+
+Ingest a meeting the moment its transcript is ready:
+
+```bash
+curl -X POST http://127.0.0.1:18789/hooks/agent \
+ -H 'Authorization: Bearer SECRET' \
+ -H 'Content-Type: application/json' \
+ -d '{"message":"A meeting transcript is ready. Pull it from the recorder MCP and write it into the meetings/ folder of this project as meetings/<source>-<source_meeting_id>.","name":"OK meeting ingest"}'
+```
+
+Prep a doc the moment a meeting starts:
+
+```bash
+curl -X POST http://127.0.0.1:18789/hooks/agent \
+ -H 'Authorization: Bearer SECRET' \
+ -H 'Content-Type: application/json' \
+ -d '{"message":"A meeting just started. Create its doc in the meetings/ folder from the prep template and link each attendee to their people/ dossier.","name":"OK meeting prep"}'
+```
+
+Which event you send matters more than the plumbing. [Ingest meetings](../workflows/meeting-ingestion.mdx) covers which recorder emits what, and why a transcript-ready event drives ingestion while `meeting.started` drives prep.
+
+One practical constraint: a vendor cannot POST to `127.0.0.1`. Keep hook endpoints behind loopback, a private network, or a trusted reverse proxy, which means a locally-run gateway needs a tunnel or a relay to receive vendor webhooks. That is why the cron poll is the default and webhooks are the upgrade. Restrict `hooks.allowedAgentIds` to limit which agents an external caller can target, and treat every inbound payload as untrusted.
+
## Preview
Write and edit responses include a route-only preview path like `/#/notes/mcp-round-trip` — a route id, **not** a full URL. Call `preview_url` for the openable browser address, and don't screenshot the preview to confirm an edit (the tool response is the confirmation). See [MCP reference → Preview links](/docs/reference/mcp#preview-links) for the full semantics.
diff --git a/docs/content/workflows/entity-vault.mdx b/docs/content/workflows/entity-vault.mdx
index cb330a280..2d65d96d7 100644
--- a/docs/content/workflows/entity-vault.mdx
+++ b/docs/content/workflows/entity-vault.mdx
@@ -124,6 +124,12 @@ Compatibility details:
- Separate compiled truth from the timeline with a `--- timeline ---` sentinel line.
- Keep timeline entries append-only and dated: `- **YYYY-MM-DD** | source | @author — event. Confidence: ...`.
+## Get meetings in from a recorder
+
+OpenKnowledge does not record meetings. Bring any recorder that speaks MCP, and each meeting lands in `meetings/` as markdown, ready for the loop below. Ask for them on demand, or have them sync automatically.
+
+See [Ingest meetings](./meeting-ingestion.mdx) for the recorder list, the `source` and `source_meeting_id` addressing rule, and the scheduled and event-driven options.
+
## Worked loop: meeting → dossiers → human correction
1. Create `meetings/2026-05-12-jane-founder-coffee.md` from the meeting template.
diff --git a/docs/content/workflows/meeting-ingestion.mdx b/docs/content/workflows/meeting-ingestion.mdx
new file mode 100644
index 000000000..f080a78bd
--- /dev/null
+++ b/docs/content/workflows/meeting-ingestion.mdx
@@ -0,0 +1,87 @@
+---
+title: Ingest meetings
+icon: LuMic
+description: "Get meeting transcripts into your knowledge base as markdown, where they are searchable, linkable, and answerable. OpenKnowledge does not record meetings. You bring a recorder, OK holds the result."
+---
+
+OpenKnowledge does not capture audio. It is the knowledge layer: the moment a meeting lands as a markdown doc it is searchable, linkable, and answerable like everything else in your knowledge base. Capture is bring-your-own, so any recorder that speaks MCP can feed OK.
+
+Meetings live in a `meetings/` folder. Any project can have one. The [Entity vault](./entity-vault.mdx) pack sets one up already and feeds each meeting into per-person dossiers.
+
+There are two ways to get meetings in:
+
+- **On demand.** You ask, the agent writes. Works in any editor, no extra infrastructure.
+- **Automatically.** A scheduler or a webhook asks for you, so meetings appear on their own. Needs the [OpenClaw](../integrations/openclaw.mdx) gateway.
+
+Both write the same doc, into the same folder, using the same addressing rule below.
+
+## Address each meeting by its source id
+
+Give every ingested meeting two frontmatter keys, and name its doc after them:
+
+```yaml
+---
+title: Roadmap sync
+date: 2026-07-08
+attendees: []
+source: granola
+source_meeting_id: abc123
+---
+```
+
+The doc lives at `meetings/<source>-<source_meeting_id>`, so the example above is `meetings/granola-abc123`.
+
+That pair is the dedup key. Because the path is derived from the recorder's own stable id, re-syncing a meeting rewrites the same doc in place instead of creating a second copy. This is what makes an unattended sync safe to run every thirty minutes.
+
+Keep the transcript verbatim. Notes and summaries are yours to edit; the transcript is the record.
+
+## On demand
+
+1. Add your recorder's MCP server to your **editor's** config, next to `open-knowledge`. OK has no MCP client of its own, so this is editor config, not OK config.
+2. Ask your agent to write recent meetings into `meetings/`.
+
+The agent pulls the transcript, notes, attendees, and date from the recorder and writes one doc per meeting through OK's `write` tool. Indexing, backlinks, and the live preview all happen on write. Nothing new to install in OK.
+
+## Which recorders work
+
+The axis that matters most is bot versus no-bot. A bot recorder joins the call as a visible participant. Granola captures locally, so nothing joins.
+
+| Recorder | Capture | MCP | Webhook |
+| --- | --- | --- | --- |
+| Fireflies | bot | official, write-capable | `Transcription completed` (HMAC-signed) |
+| Circleback | bot | official | via Zapier |
+| tl;dv | bot | official (self-host) | yes |
+| Fathom | bot | official | via integrations |
+| Fellow | bot | official | via Zapier |
+| Granola | local, no bot | official, read-only | none native, bridges via Zapier |
+
+Otter, Avoma, Teams (MS Graph), and Google Meet expose an API but no official MCP yet, so they need more wiring. Verify a recorder's MCP endpoint is live before relying on it. A few of these are asserted on vendor launch pages rather than deep developer docs.
+
+## Automatically, on a schedule
+
+The [OpenClaw](../integrations/openclaw.mdx) gateway has a built-in scheduler. Register both MCP servers with it, `open-knowledge` via `ok init` and your recorder's MCP per the vendor's install guide, then create one cron job that wakes an agent to run the on-demand flow for you:
+
+```bash
+openclaw cron create "*/30 * * * *" \
+ "Pull meetings from the recorder MCP since the last run. Write each new one into the meetings/ folder of this project, addressed as meetings/<source>-<source_meeting_id> so re-runs update in place. Do not duplicate meetings already present." \
+ --name "OK meeting sync" \
+ --session isolated \
+ --no-deliver
+```
+
+Polling works for every recorder, including Granola, which has no native webhook. The gateway must be running for schedules to fire.
+
+## Automatically, on an event
+
+If you want lower latency than a poll, the gateway can run the same agent the moment something happens. The wiring (hooks config, endpoints, auth) lives in [trigger runs from external events](../integrations/openclaw.mdx#trigger-runs-from-external-events).
+
+Choosing which event to send matters more than the plumbing, because the two moments are different jobs:
+
+| Event | Fires when | Use it for |
+| --- | --- | --- |
+| `Transcription completed` (Fireflies), note added (Granola via Zapier) | the transcript is ready, a few minutes after the meeting ends | **Ingestion.** There is nothing to write before this. |
+| `meeting.started` (Zoom) | the moment the meeting begins | **Prep.** Create the doc and link attendees before anyone speaks. |
+
+Google Calendar push notifications only signal that the calendar changed and carry no event data, so they need a follow-up API call before they are useful here.
+
+One practical constraint: a vendor cannot POST to `127.0.0.1`. A locally-run gateway needs a tunnel or a relay to receive vendor webhooks, which is why the cron poll is the default and webhooks are the upgrade.
diff --git a/docs/content/workflows/meta.json b/docs/content/workflows/meta.json
index e26d7de20..e3f482034 100644
--- a/docs/content/workflows/meta.json
+++ b/docs/content/workflows/meta.json
@@ -9,6 +9,7 @@
"plain-notes",
"worldbuilding",
"supporting-open-knowledge-format",
+ "meeting-ingestion",
"entity-vault"
]
}
diff --git a/packages/server/assets/skills/packs/entity-vault/SKILL.md b/packages/server/assets/skills/packs/entity-vault/SKILL.md
index 7bd2aff9e..038626701 100644
--- a/packages/server/assets/skills/packs/entity-vault/SKILL.md
+++ b/packages/server/assets/skills/packs/entity-vault/SKILL.md
@@ -26,7 +26,7 @@ When a new fact arrives, route it: update **compiled truth** if it changes curre
## Folders
- **`people/`**, **`companies/`**, **`concepts/`** — dossiers (compiled truth + timeline). Frontmatter `type: person|company|concept`.
-- **`meetings/`** — meeting notes (`YYYY-MM-DD-<slug>.md`); `attendees:` should match dossier filenames in `people/`. The verbatim record — do NOT rewrite it.
+- **`meetings/`** — meeting notes (`YYYY-MM-DD-<slug>.md`); `attendees:` should match dossier filenames in `people/`. The verbatim record — do NOT rewrite it. When a note is ingested from a meeting-recorder tool, add `source:` (e.g. `granola`, `fireflies`) and `source_meeting_id:` (the recorder's stable id) to its frontmatter — those two are the dedup key.
- **`originals/`** — your own untransformed thinking; authoritative source material. Frontmatter `type: original`.
- **`media/`** — bulk transcripts, voice notes, large attachments; often `.okignore`-d to keep the index light.
@@ -41,6 +41,7 @@ Two deltas from the platform default, both GBrain-specific:
## Agent behaviors
+- **Ingest from a meeting recorder.** If a meeting-recorder MCP (Granola, Fireflies, Circleback, tl;dv, Fathom, …) is registered alongside `open-knowledge`, pull recent meetings and write each into `meetings/`. Address each as `meetings/<source>-<source_meeting_id>` so re-syncing a meeting updates it in place instead of creating a duplicate. Normalize whatever the recorder returns (Granola's Me/Them streams, a bot's speaker turns) into notes plus the verbatim transcript; do NOT rewrite the transcript. This works the same on demand (you ask) or unattended (a scheduler asks) — the recorder is bring-your-own and interchangeable.
- After a meeting note lands, extract entity mentions and append timeline bullets to each referenced dossier (cite the meeting by markdown link). Stub any mentioned entity not yet captured.
- Treat `originals/` as authoritative (the user's own words, not inferences).
- Surface entity-to-entity edges (person ↔ company, concept hubs) when both ends exist.
diff --git a/packages/server/src/seed/starter.test.ts b/packages/server/src/seed/starter.test.ts
index 822a82831..06cecf596 100644
--- a/packages/server/src/seed/starter.test.ts
+++ b/packages/server/src/seed/starter.test.ts
@@ -431,6 +431,14 @@ describe('Entity vault pack — GBrain-compatible Markdown shape', () => {
'[[concepts/agent-runtime-observability|agent-runtime observability]]',
);
});
+
+ test('the meeting template carries the recorder dedup key and a verbatim transcript section', () => {
+ // `meetings/<source>-<source_meeting_id>` is the doc path, so re-syncing a
+ // meeting rewrites it in place instead of creating a duplicate.
+ expect(ENTITY_VAULT_PACK.templates.meeting).toContain('source:');
+ expect(ENTITY_VAULT_PACK.templates.meeting).toContain('source_meeting_id:');
+ expect(ENTITY_VAULT_PACK.templates.meeting).toContain('## Transcript');
+ });
});
describe('Codebase wiki pack — nested wiki/ layout', () => {
diff --git a/packages/server/src/seed/starter.ts b/packages/server/src/seed/starter.ts
index 1936337a4..e0f342f25 100644
--- a/packages/server/src/seed/starter.ts
+++ b/packages/server/src/seed/starter.ts
@@ -1162,7 +1162,7 @@ const ENTITY_VAULT_FOLDERS: readonly StarterFolder[] = [
title: 'Meetings',
uiSummary: 'Notes from your meetings.',
description:
- 'Meeting notes. Filename `YYYY-MM-DD-<slug>.md`. Frontmatter carries `title`, `date`, `attendees:` (prefer person slugs/names that resolve to `people/` dossiers), and `type: meeting`. Body is raw notes with path-qualified links to the people, companies, and concepts mentioned. Agent: after a meeting note lands, extract entity mentions and append dated timeline bullets to each referenced dossier. Do NOT rewrite the meeting note; it is the verbatim record.',
+ 'Meeting notes. Filename `YYYY-MM-DD-<slug>.md`. Frontmatter carries `title`, `date`, `attendees:` (prefer person slugs/names that resolve to `people/` dossiers), and `type: meeting`. A meeting ingested from a recorder also carries `source` and `source_meeting_id`; its doc path `meetings/<source>-<source_meeting_id>` derives from that pair, so re-syncing the same meeting updates it in place instead of creating a duplicate. Body is raw notes with path-qualified links to the people, companies, and concepts mentioned. Agent: after a meeting note lands, extract entity mentions and append dated timeline bullets to each referenced dossier. Do NOT rewrite the meeting note; it is the verbatim record.',
tags: ['meeting', 'note'],
starterTemplate: 'meeting',
},
@@ -1249,6 +1249,8 @@ description: "Meeting notes: attendees, discussion, decisions, and action items.
title: Meeting Title
date: {{date}}
attendees: []
+source:
+source_meeting_id:
author: {{user}}
tags: [meeting]
---
@@ -1257,6 +1259,10 @@ tags: [meeting]
(Raw notes from the meeting. Prefer path-qualified links such as \`[[people/jane-founder|Jane Founder]]\`, \`[[companies/jane-co|Jane Co]]\`, and \`[[concepts/agent-runtime-observability|agent-runtime observability]]\`.)
+## Transcript
+
+(Verbatim transcript, when the meeting was ingested from a recorder. Leave empty for notes you typed yourself. Never rewrite it.)
+
## Action items
- [ ]