by memaraio · v0.1.5
Give Dify agents durable long-term memory in Memara — create, semantic search, list, fetch, update, and delete memories with your Memara Dify API key (sk_dify_*).
This community listing does not yet include every recommended support, privacy, pricing, and permission disclosure. Review the available package permissions before installing.
Available inside your emploidai workspace after installation.
Available inside your emploidai workspace after installation.
Memara is long-term memory for AI: durable facts and context your agents can search, update, and reuse. This official Dify tool plugin connects Dify workflows and agents to your Memara workspace using a dedicated API key.
Maintainers: development, marketplace PRs, and release automation are documented in MAINTAINER.md.
sk_dify_*) and Memara’s HTTP API (/api/v1/dify).sk_dify_… API key.https://api.memara.io) — origin only, without /api/v1.From the Dify marketplace (when your version is listed): install Memara like any other tool plugin.
From a release package: open github.com/memaraio/memara-dify-plugin/releases, download the .difypkg for your version, then in Dify use Plugins → Install from package file (wording may vary slightly by Dify version).
Self-hosted Dify: run your own Dify stack and install Memara from a local .difypkg (bypasses Cloud marketplace issues). See SELF_HOSTED_DIFY.md for a bootstrap script, manual deploy links, Memara install steps, and optional remote debug against your plugin_daemon.
Registry / marketplace updates ship a versioned .difypkg inside the PR to langgenius/dify-plugins (not as a file on this repo’s main); see MAINTAINER.md.
In Dify, open the Memara provider settings and set:
| Field | Value |
|---|---|
| API base URL | Your Memara API origin, e.g. https://api.memara.io |
| API key | Your sk_dify_… key (Memara treats it as a Bearer token) |
Saving runs a validation call to your integration (GET /api/v1/dify/integration/info).
| Tool | Purpose |
|---|---|
| Create memory | Store new memory content (and optional metadata). |
| Search memories | Semantic search over your memories. |
| List memories | Paginated list (good for dashboards and polling). |
| Get memory | Fetch one memory by id. |
| Update memory | Change fields on an existing memory. |
| Delete memory | Remove a memory by id. |
offset must be ≥ 0.To refresh on a timer, add a Schedule (or interval) trigger in your workflow, then call List memories and/or Search memories, passing variables such as last-seen created_at or a cursor from the previous run.