Memory
Adding sources
Supported source types, indexing states and re-indexing.
Availability note — these docs describe both working and planned product areas. Account authentication, administrator-managed AIHK/AICA tenant regions and keys, signed-in Composer runs and public Guard scans are connected. Skills, Memory, Agents, Workflows, Apps, billing, the public SDK/CLI and the unified API are previews unless a section explicitly says otherwise.
Roadmap and package-contract reference: this page describes the intended GenCMS runtime. The current deployment persists workspace records and drafts but does not execute marketplace Skills, index Memory for retrieval, publish packages, or run Agents, Workflows or Apps. See the deployed API reference for the routes available today.
| Source | Accepted | Notes |
|---|---|---|
| Note | Pasted text | Fastest way to add a rule or a decision |
| Document | PDF, DOCX, Markdown, TXT | Text is extracted; page anchors are kept for citation |
| URL | Single page | Re-fetched on a schedule you set |
| Website | Crawled section | Bounded by path prefix and page count |
| Repository | Git | Indexed by file, respecting the ignore file |
| Structured data | CSV, JSON | Kept as records rather than prose |
| API documentation | OpenAPI, Markdown | Endpoints indexed individually |
Indexing states
- queued — accepted, waiting for a worker
- extracting — text and structure being pulled out
- chunking — split into retrievable units with anchors
- embedding — vectors written
- indexed — retrievable, with a last-indexed timestamp
- failed — with the reason, and a retry that does not duplicate the source
A source that fails stays visible with its error. Silent partial indexes are worse than an obvious failure, because retrieval then returns confidently incomplete answers.