fix(openai-sdk-python): migrate to v4 Supermemory APIs

[Dhravya]

Summary

• Replace deprecated search.execute with search.memories (hybrid mode) in search_memories
• Replace memories.add with client.add in tools and middleware
• Fix middleware container_tag param (was incorrectly container_tags)
• Fix profile memory deduplication for string and Pydantic API items
• Bump supermemory>=3.50 and requires-python>=3.9

Test plan

• uv run pytest tests/test_tools.py::TestMemoryOperationsUnit

Made with Cursor

[claude]

Claude finished @Dhravya's task in 2m 33s —— View job

---

Reviewing PR #1429: migrate openai-sdk-python to v4 Supermemory APIs

• Fetch and analyze PR diff
• Review existing comments to avoid duplicates
• Read changed files for context
• Verify reported issues and check for additional bugs
• Submit review findings

Summary

Reviewed the migration from v3 to v4 Supermemory APIs. Confirmed 2 issues that need to be addressed before merge:

1. container_tags vs container_tag — The middleware was correctly fixed to use container_tag (singular), but tools.py still passes container_tags (plural) at lines 188 and 223. This inconsistency could cause API failures.

2. include_full_docs silently ignored — The parameter is still accepted but no longer forwarded to the API, which is a breaking behavior change.

Additionally noted that the unit tests validate the current (incorrect) implementation and will need updating.

Score: 7/10 — Good migration overall, but the issues above need fixing before merge.

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
✅ Deployment successful! View logs	supermemory-mcp	d348090	Aug 08 2026, 03:00 AM

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	supermemory-app	d348090	Commit Preview URL Branch Preview URL	Aug 08 2026, 03:01 AM

[claude]

Overview: Migrates the openai-sdk-python package from deprecated v3 Supermemory APIs to v4, updating search and add operations.

Issues found:

• include_full_docs parameter in search_memories() is accepted but silently ignored after the migration — callers relying on this will get unexpected behavior.

The rest of the changes look solid:

• Correct migration from search.execute to search.memories with hybrid mode
• Proper fix of container_tags → container_tag in middleware (passing single value vs list)
• Good defensive handling with getattr(response, "id", None) for potentially missing attributes
• Utils deduplication correctly extended to handle Pydantic models alongside dicts and strings
• Unit tests cover the new API contract

Score: 9/10

Minor issue that should be addressed before merge to avoid silent behavior change for existing callers.

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 2 potential issues.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 03849ce. Configure here.}

[Dhravya]

• feat(sdk-playground): interactive SDK chat playground #1437
• chore(ci): wire JS tools unit tests into CI #1436
• docs(skills): refresh Supermemory skill for 7-tool parity #1435
• fix(python-sdks): v4 API migration for integration packages #1434
• feat(ai-sdk): re-export 7-tool surface #1433
• feat(tools): 7-tool parity and description refresh #1432
• chore(ci): Python SDK pytest workflow #1431
• feat(openai-sdk-python): 7-tool parity #1430
• fix(openai-sdk-python): migrate to v4 Supermemory APIs #1429 👈 (View in Graphite)
• main

---

How to use the Graphite Merge Queue

Add the label Main to this PR to add it to the merge queue.

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has enabled the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

This stack of pull requests is managed by Graphite. Learn more about stacking.

[claude]

Overview: Migrates the openai-sdk-python package from deprecated v3 Supermemory APIs to v4, updating search and add operations.

Issues found:

1. container_tags vs container_tag inconsistency (High): The middleware was correctly fixed to use container_tag (singular) at line 227, but tools.py still passes container_tags (plural) at lines 188 and 223. If the v4 API only accepts container_tag, these calls will fail or silently ignore the tenant scope.

2. include_full_docs silently ignored (Medium): The search_memories function still accepts include_full_docs in its signature (line 172) and tool schema (lines 68-75), but the v4 client.search.memories call no longer forwards it. Callers relying on this behavior will get unexpected results.

3. Unit tests assert incorrect behavior: The tests in test_tools.py (lines 202-205, 227) assert that container_tags is passed, which validates the current (incorrect) implementation rather than the correct v4 API contract. These tests will need updating when the code is fixed.

What looks good:

• Correct migration from search.execute to search.memories with hybrid mode
• Proper defensive handling with getattr(response, "id", None) for potentially missing attributes
• Utils deduplication correctly extended to handle Pydantic models alongside dicts and strings
• Good null-safety with response.results or []

Score: 7/10

The two issues above need to be addressed before merge — the container_tags bug could cause silent failures in production, and the include_full_docs change is a breaking behavior change for existing callers.

[cursor]

Breaking changes (v2.0) — please fold into PR description

Package bumped 1.0.4 → 2.0.0 (semver major for public config API break).

SupermemoryToolsConfig.container_tags (list) is removed. Use a single container_tag string instead. Search/add tool calls now send singular container_tag to the Supermemory v4 API.

# Before (v1)
config = {"container_tags": ["user-123"]}

# After (v2)
config = {"container_tag": "user-123"}

Also in this release:

• include_full_docs → include={"documents": ...} on client.search.memories
• uv.lock, Trove classifiers, and mypy aligned with requires-python>=3.9