--- title: "Glean to Slack Canvas Migration | ClonePartner" description: "Glean to Slack Canvas: Glean's API architecture is strictly inbound-only; the Indexing API pushes documents into Glean but there is no corresponding endpoint…" source: Glean target: Slack Canvas canonical: "https://clonepartner.com/migrate/glean-to-slack-canvas" --- # Glean → Slack Canvas migration Glean to Slack Canvas migration — documents moved in bulk and attached to the channels your team already works in. ## Why teams migrate from Glean to Slack Canvas Migrating from Glean to Slack Canvas is not a straightforward platform-to-platform transfer because Glean functions as an enterprise search and AI indexing layer rather than a content management system. No native export path exists: Glean's Indexing API is strictly inbound-only, the Client API returns metadata and snippets rather than full document bodies, and there is no tenant-wide data download endpoint. The actual migration effort splits into two workstreams: extracting Glean-native objects (Collections, Answers, Shortcuts, Announcements, and Pins) via the Client API and mapping them into Slack canvases, plus separately migrating upstream source content from the platforms Glean was indexing such as Confluence, Google Drive, and SharePoint. Custom scripting is required for extraction, pagination handling, and canvas creation, with additional complexity around Announcement objects that have no list endpoint in the public API. ## Key migration challenges - **No Export Endpoint Exists** — Glean's API architecture is strictly inbound-only; the Indexing API pushes documents into Glean but there is no corresponding endpoint to retrieve full indexed document bodies or perform a tenant-wide export. - **Announcements Lack List Endpoint** — The Glean Client API exposes only create, update, and delete operations for Announcements, with no list or read-all endpoint, requiring ID recovery through browser DevTools inspection, manual logging, or a Glean support escalation. - **Inconsistent API Scope Breadth** — List endpoints for Collections and Pins return org-wide results, while Answers and Shortcuts return only objects owned by the authenticated user, requiring owner-by-owner iteration or delegated auth for complete tenant extraction. - **Pagination Truncation Risk** — All Glean list endpoints use cursor-based pagination on large tenants, and extraction scripts that omit cursor handling will silently return only the first page with no error or warning. - **Upstream Source Content Out of Scope** — The vast majority of content surfaced through Glean — Confluence pages, Google Drive files, SharePoint sites — lives in external platforms that each require their own separate migration workstream independent of Glean. - **Canvas Type Selection Complexity** — Slack provides two canvas types with different access models (channel canvases tied to channel membership versus standalone canvases requiring explicit access grants), and selecting the wrong type for a given Glean object type will require rework. ## Entity mappings - **Document** (Glean) → **Standalone Canvas** (Slack Canvas) - **Collection** (Glean) → **Canvas Content (Markdown)** (Slack Canvas) - **Answer** (Glean) → **Standalone Canvas** (Slack Canvas) - **Announcement** (Glean) → **Channel Canvas** (Slack Canvas) - **Pin** (Glean) → **File Attachment** (Slack Canvas) - **Shortcut** (Glean) → **Canvas Content (Markdown)** (Slack Canvas) - **Team** (Glean) → **Channel** (Slack Canvas) - **Group** (Glean) → **Canvas Access Grant** (Slack Canvas) - **User** (Glean) → **Canvas Access Grant** (Slack Canvas) - **ChatMessage** (Glean) → **Canvas Content (Markdown)** (Slack Canvas) - **SearchQuery** (Glean) → **NotAvailable** (Slack Canvas) - **InsightsReport** (Glean) → **Slack Canvas (Document)** (Slack Canvas) - **DatasourceSchema** (Glean) → **Table** (Slack Canvas) ## What breaks during migration - **Collections** [Direct] — Collections have a supported org-wide list endpoint and map nearly 1:1 to Slack index canvases containing categorized links, making them the most straightforward object to extract and migrate. - **Answers** [Workaround] — The Answers list endpoint returns only objects owned by the authenticated user rather than the full org-wide set, requiring owner-by-owner extraction or delegated authentication to achieve completeness. - **Shortcuts (Go Links)** [Workaround] — Like Answers, Shortcuts are scoped to the authenticated user's owned or editable objects, preventing a single clean tenant-wide pull and requiring additional coordination to extract all alias-to-URL mappings. - **Announcements** [Breaks] — Announcements have no list or read-all endpoint in the public API, meaning ID recovery depends on browser DevTools inspection of admin console traffic, proactive manual logging, or a Glean support escalation — creating a high risk of data loss for ad hoc announcements created without ID logging. - **Pins** [Breaks] — Pins are a search-layer concept that reference documents by ID for specific queries, and Slack has no equivalent 'pin to search' feature, meaning this object type has no direct migration target and its functionality cannot be replicated in Slack Canvas. - **Indexed Document Bodies** [Breaks] — Full document body text indexed from external sources cannot be retrieved from Glean via any API endpoint; the Client API search endpoint returns only titles, snippets, and URLs, making upstream source platforms the required migration origin for this content. - **Slack Message Content** [Breaks] — Glean does not persist Slack message bodies in its index — messages are retrieved live from Slack via the RTS API at query time and processed in memory only — so there is no Slack message content stored in Glean to migrate. - **API Authentication Tokens** [Workaround] — Using a non-admin Glean Client API token will successfully authenticate but silently return only user-owned objects rather than the org-wide set, creating a high risk of incomplete extraction if token scope is not validated before running the migration script. - **Canvas Access Permissions** [Workaround] — Standalone Slack canvases are visible to no one until access is explicitly granted via canvases.access.set, meaning migrated content will be inaccessible if the permission-granting step is omitted from the migration workflow. - **Upstream Source Content (Confluence, Drive, SharePoint)** [Breaks] — Tens of thousands of documents referenced by Glean Collections and Answers live in external source platforms and represent a separate, months-long migration workstream that is entirely outside the scope of the Glean API extraction process. ## What we migrate ### Migration Filter - **Time Range** — Migrate articles created or updated within a specific period - **Language & Locale** — Migrate specific language versions or translations only - **Category Selection** — Select specific sections, folders, or categories to migrate - **Publication Status** — Filter by published, draft, archived, or internal-only articles ### Data Types - **Tags & Metadata** — SEO settings, search labels, author attribution, and article tags - **Articles & Content** — Core help articles including HTML formatting and inline images - **Media & Attachments** — All downloadable files, PDFs, and media assets embedded in articles - **Categories & Hierarchy** — Full folder structure with sections, sub-sections, and parent categories ## Complete technical guide For a deep-dive into the technical process, data mapping, and step-by-step migration workflow, read our full guide: [Glean to Slack Canvas Migration: Why the Export Doesn't Exist](https://clonepartner.com/blog/glean-to-slack-canvas-migration-why-the-export-doesnt-exist/). ## Frequently asked questions ### Can you export data from Glean? No. Glean's Indexing API is inbound-only, and the Client and Platform APIs are query-based. There is no bulk export, tenant export, or full document retrieval endpoint. You can extract Glean-native objects like Collections, Answers, Shortcuts, and Pins via the Client API, but indexed documents must be exported from their original source platforms. ### What content is actually stored in Glean? Glean stores search index data, metadata, and references to documents from connected sources. The only content created and stored natively in Glean includes Collections (curated document lists), Answers (FAQ pairs), Shortcuts (go-links), Announcements, and Pins. All other content lives in upstream platforms like Confluence, Google Drive, or SharePoint. ### What are the Slack Canvas API rate limits for migration? canvases.create and conversations.canvases.create are Tier 2 (20+ per minute). canvases.edit and canvases.access.set are Tier 3 (50+ per minute). Canvas markdown content is limited to 1 MiB per document_content object, and tables are capped at 300 cells. ### Do Glean permissions transfer to Slack Canvas? No. Glean mirrors source-system permissions at query time, while Slack canvas access comes from channel membership or explicit channel/user sharing via canvases.access.set. You must rebuild permissions from scratch based on your Slack channel structure. ### Does Glean index Slack canvases after migration? Glean's Slack RTS connector requests canvases:read and canvases:write scopes, which confirms it is canvas-aware. However, Glean's public docs emphasize message search, not canvases as a documented indexed content type. Verify behavior in your own tenant before assuming canvases are searchable in Glean. ## Get a fixed-price quote [Talk to an engineer](https://clonepartner.com/talk-to-us?duration=30&utm_source=xtoy&utm_medium=button&utm_campaign=demo_bookings&utm_content=cta_click&utm_term=demo_button_click) about your Glean → Slack Canvas migration.