docs(skills): refresh mem0 skills against SDK 2.2.1 / TS 3.3.1 (#7548)

skills/mem0-integrate/SKILL.md

@@ -15,7 +15,7 @@ description: >
 license: Apache-2.0
 metadata:
   author: mem0ai
-  version: "0.1.0"
+  version: "0.1.1"
   category: ai-memory
   tags: "memory, integration, tdd, platform, oss"
   mem0_tested_versions: "mem0ai (PyPI) >=2.0.0,<3.0.0; mem0ai (npm) >=3.0.0,<4.0.0"
@@ -36,7 +36,7 @@ of the Mem0 API.
 - Scope-tagged docs index: https://docs.mem0.ai/llms.txt
 - Full docs (single file, deep dives): https://docs.mem0.ai/llms-full.txt
 - OpenAPI spec (Platform REST, machine-readable): https://docs.mem0.ai/openapi.json
-- Hosted MCP server: https://mcp.mem0.ai (requires Platform API key)
+- Hosted MCP server: https://mcp.mem0.ai/mcp (requires Platform API key)
 - Integrations index: https://docs.mem0.ai/integrations
 
 ### Published Mem0 skills — delegate; do not reimplement
@@ -109,7 +109,7 @@ the target stack. If yes, delegate — copy its call-site pattern into
 |---|---|---|
 | `@ai-sdk/*` + `ai` in `package.json` | `skills/mem0-vercel-ai-sdk` | Integration is via `createMem0` provider wrapper, not raw `MemoryClient`. |
 | CLI-only repo (Typer, Commander, Click, Cobra) with no LLM call sites | `skills/mem0-cli` | Call sites are command handlers, not model wrappers. Consider whether mem0 actually fits first. |
-| Target is an MCP client / editor config (Claude Code, Cursor, Codex settings) | `integrations/mem0-agent-plugin` | Wire via MCP server URL + hooks; no SDK code usually needed. |
+| Target is an MCP client / editor config (Claude Code, Cursor, Codex settings) | `integrations/mem0-agent-plugin` | Local stdio MCP server (one search tool) plus skills, no hooks or automatic capture; for a hosted endpoint use `https://mcp.mem0.ai/mcp`. No SDK code usually needed. |
 | Any other Python or TS repo with an LLM call site | `skills/mem0` | Default SDK integration path. |
 
 Record the delegated skill's raw URL in `plan.md` under a
@@ -168,7 +168,7 @@ start executing a step; the summary below is only for routing.
 | `trace.jsonl` | Every tool call, decision, and subagent exchange this run. | Overwritten per run. |
 | `diff.patch` | The committed integration as a reviewable patch. | Overwritten per run. |
 | `heal-trace.md` | Per-attempt record of the self-healing loop (step 10). | Overwritten per run. |
-| `product.json` | `{"product": "platform"\|"oss", "language": "...", "mem0_version": "...", "write_site": "file:line", "read_site": "file:line", "feature_flag": "MEM0_ENABLED"}` — consumed by the verification skill. | Overwritten per run. |
+| `product.json` | `{"product": "platform"\|"oss", "language": "...", "mem0_version": "...", "write_site": "file:line", "read_site": "file:line", "feature_flag": "MEM0_ENABLED", "preferred_site": "<surface index from step 2>"}`, consumed by the verification skill. | Overwritten per run. |
 
 `.mem0-integration/` is added to `.gitignore` on first run. Nothing is
 written outside this directory and the repo's source tree.

skills/mem0-integrate/references/pipeline.md

@@ -116,8 +116,11 @@ Present in env, continue.
 Mode**: run `mem0 init --agent --agent-caller <your-name> --json` (after
 `pip install mem0-cli` or `npm install -g @mem0/cli`), substituting your agent
 identity such as `claude-code`, `cursor`, `codex`. If you forgot
-`--agent-caller`, run `mem0 identify <your-name>` after init. Cache the key to
-`.env` with user consent and continue. Tell the user to claim it later with
+`--agent-caller`, run `mem0 identify <your-name>` after init. Init does not
+print the key: it saves it to `~/.mem0/config.json` (`platform.api_key`, file
+mode 0600) and reuses a valid key already there instead of minting a new one,
+so read the key from that file. Cache the key to `.env` with user consent and
+continue. Tell the user to claim it later with
 `mem0 init --email <their-email>`: same key, no agent disruption.
 
 Missing and **CI mode** (`MEM0_INTEGRATE_CI=1`), exit code 2 with the name of
@@ -184,8 +187,8 @@ Then write `.mem0-integration/plan.md`:
     call client.add([user_msg, assistant_msg], user_id=<source>).">
 
     **Read pattern:** <one sentence, e.g. "Before building the LLM prompt,
-    call client.search(query=latest_user_msg, user_id=<source>, limit=5)
-    and inject results as a system message.">
+    call client.search(latest_user_msg, filters={"user_id": <source>},
+    top_k=5) and inject results["results"] as a system message.">
 
     **User identifier source:** <code path, e.g. `req.auth.userId`,
     `session.user.email`, `ctx.params.user_id`. If none, ask the user.>
@@ -276,7 +279,8 @@ test framework:
 Test assertion shapes must match the **canonical signatures**:
 
 - Platform method signatures: `https://docs.mem0.ai/openapi.json`, the request
-  body schemas for `/v1/memories/` and `/v1/memories/search/`.
+  body schemas for `/v3/memories/add/` and `/v3/memories/search/` (the
+  `/v1/memories/` POST and `/v1/memories/search/` paths are deprecated).
 - OSS method signatures: the delegated skill named in `plan.md` (fetched from
   its raw URL), or `skills/mem0/SKILL.md` as the default.
 - Do not hand-roll request shapes. If the delegated skill has an example
@@ -341,7 +345,8 @@ Otherwise loop:
 
 1. **Categorize the failing check** from `scorecard.json` and route:
    - `install` / `static_checks`, dependency or import fix.
-   - `unit_tests`, wiring or assertion fix.
+   - `unit_tests_flag_on`, wiring or assertion fix. (`unit_tests_flag_off`
+     failing is the non-invasiveness case below.)
    - `smoke_test`, API key or SDK call-shape fix.
    - `e2e_test`, recipe, flag-wiring, or integration-point fix.
    - **Pre-existing test failure** (test skill exit code 7,
@@ -351,7 +356,7 @@ Otherwise loop:
 
 2. **Spawn a remediation subagent** with fresh context. Inputs: `plan.md`,
    `goal.md`, `scorecard.md`, `scorecard.json`, the last committed diff, and
-   the relevant log for the failing category (`test-stdout.log` /
+   the relevant log for the failing category (`test-stdout-flag-on.log` /
    `smoke-stdout.log` / `e2e-app.log` / `e2e-calls.log`). Use the remediation
    prompt in [`subagent-prompts.md`](subagent-prompts.md) verbatim.
 

skills/mem0-integrate/references/subagent-prompts.md

@@ -31,7 +31,8 @@ so paraphrasing them drops constraints the review step then has to catch.
     6. Preserve everything listed under plan.md's "Preserved behavior"
        and "Coexistence."
     7. Lazy client construction. `MemoryClient()` validates the API
-       key in `__init__` (it makes a network call). Never instantiate
+       key in `__init__` (Python pings the API; TS throws on a missing
+       or blank key and pings in the background). Never instantiate
        it at module-import time, construct on first use inside the
        request / handler path. The same rule applies to OSS `Memory()`,
        which can eagerly initialize embedding and LLM providers. Use