AI coding assistants

Ask a coding assistant to build a Membase integration, with checks for learning completion, permissions and search failures.

Use this recipe when a coding assistant is implementing Membase in your application. To give the coding assistant your memory for its own work, follow its client guide under Connect your AI instead.

Install the Membase skill using your client’s setup guide (Claude Code, Codex, Cursor, Kimi Code). Then give it the integration task below.

Let the agent build the integration

The same skill carries the API reference, the SDK guide, the MCP setup and the use cases, so an agent that has it can write the integration for you. Install it once, then prompt:

Add Membase memory to this app. Read the membase skill first. Use the SDK; read the profile once per session when granted, search before answering about past work, and save only what the user supplied. Wait for learning before searching newly added documents and check containers[].error. Never pass confirm on your own. Keep the key in the environment.

What the skill teaches, so you can check the result:

The agent should	Because
read get_profile once, near the start, when granted	it is small and standing; the memory of who the user is
call search_memories before answering about past work, decisions or preferences	search is retrieval inside the user’s own container
cite container_name	every hit says which Memory it came from
use add_memory for a fact, add_document for text or a URL, with custom_id	documents are raw material until the Memory’s run; the same custom_id twice is a no-op
pass confirm=true only after the person agreed	on a Full access key, a delete or forget without it answers with a how sentence, not an action
treat 403 as withdrawn access and stop	the owner narrowed or revoked the key on Connect
use a suitable timeout and inspect containers[].error	a cold runtime or failed search must not be reported as an empty Memory

The skill is served at https://www.app.membase.io/skill (its SKILL.md with the install preamble) and as a zip at https://www.app.membase.io/plugin/membase-skill.zip. Its files are under https://www.app.membase.io/plugin/skills/membase/: SKILL.md, README.md and references/api-reference.md, references/sdk-guide.md, references/mcp-setup.md, references/quickstart.md, references/use-cases.md. That directory address itself is not a listing; fetch the files by name, or take the zip.

Without the skill

Any agent can be handed the OpenAPI document instead and asked to generate a client:

curl -s https://www.app.membase.io/plugin/openapi/agent-protocol.json -o agent-protocol.json
npx openapi-typescript agent-protocol.json -o membase.d.ts

Three requests cover most integrations: GET /v1/profile, POST /v1/search with {"q": …} and POST /v1/memories with {"content": …}, each with the bearer header. API reference has every operation.