fix(mem0): scope prefetch search and result cache by user_id

[dizhaky]

Summary

• queue_prefetch() was ignoring its user_id argument and searching under self._user_id (set once at initialize() time). In shared gateway sessions where the cached AIAgent is reused across participants, this caused cross-user memory leakage.
• prefetch() returned cached results without checking which user they were fetched for, so a result queued for User A could be injected into User B's next prompt.

Changes

• queue_prefetch(): capture effective_user_id = user_id or self._user_id before spawning the background thread; pass {"user_id": effective_user_id} directly to client.search() instead of self._read_filters(); store result as (user_id, text) tuple.
• prefetch(): discard the cached result when its owner user_id does not match the current caller — prevents stale cross-user results from leaking into the wrong session.
• __init__: initialize _prefetch_result to None instead of "" to match the new tuple-or-None shape.
• Bugbot fix: empty-text tuples ((user_id, "")) now return "" rather than injecting a bare ## Mem0 Memory header with no bullets.

Addresses Codex PR#33 P1 review comment: Scope prefetched memory by current user.

All CI checks passed on the original draft PR #43.

---

Generated by Claude Code

---

Note

Medium Risk
Touches per-turn memory injection in multi-user gateway paths; behavior change is intentional isolation with limited surface area in one provider.

Overview
Fixes cross-user memory leakage when a shared Mem0MemoryProvider serves multiple gateway participants: background prefetch now searches and caches under the current turn’s user_id, not only self._user_id from initialize().

queue_prefetch() resolves effective_user_id = user_id or self._user_id, passes {"user_id": effective_user_id} to client.search() (instead of self._read_filters()), and stores the cache as (user_id, bullet text).

prefetch() returns cached bullets only when the tuple’s user matches the caller; mismatches or empty text yield "" so User A’s prefetch cannot appear in User B’s prompt and empty results do not inject a bare ## Mem0 Memory header.

__init__ sets _prefetch_result to None for the tuple-or-None cache shape.

^{Reviewed by Cursor Bugbot for commit 3a3d3ce. Configure here.}

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[github-actions]

🔎 Lint report: fix/mem0-prefetch-user-scoping vs origin/main

ruff

Total: 0 on HEAD, 0 on base (➖ 0)

🆕 New issues: none

✅ Fixed issues: none

Unchanged: 0 pre-existing issues carried over.

ty (type checker)

Total: 8649 on HEAD, 8649 on base (➖ 0)

🆕 New issues: none

✅ Fixed issues: none

Unchanged: 4570 pre-existing issues carried over.

Diagnostics are surfaced as warnings — this check never fails the build.