# Cursor SDK failing "\[unknown\] \[internal\] Protocol error" (~15–20s)

**URL:** <https://forum.cursor.com/t/cursor-sdk-failing-unknown-internal-protocol-error-15-20s/172083>\
**Category:** Bug Reports\
**Tags:** cursor-sdk, networking\
**Created:** [September 18, 2026, 12:28am UTC](https://forum.cursor.com/t/cursor-sdk-failing-unknown-internal-protocol-error-15-20s/172083 "2026-09-18T00:28:10Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Krishna\_Ponnamineni](https://avatars.discourse-cdn.com/v4/letter/k/eb9ed0/32.png) [@Krishna\_Ponnamineni](https://forum.cursor.com/u/Krishna_Ponnamineni)\
**Post date:** [September 18, 2026, 12:28am UTC](https://forum.cursor.com/t/cursor-sdk-failing-unknown-internal-protocol-error-15-20s/172083/1 "2026-09-18T00:28:10Z")

</div>

### Where does the bug appear (feature/product)?

Cursor SDK

### Describe the Bug

## Summary

Headless agents started via the Python `@cursor/sdk` (`cursor-sdk==1.0.31`) consistently fail after session start:

1. `Agent.create(...)` succeeds
2. `agent.send(...)` returns a run that emits `status=RUNNING`
3. ~15–20 seconds later: `status=ERROR` with message **`[unknown] [internal] Protocol error`**
4. No assistant text, no tool calls, empty usage / empty result

Auth and model catalog work. Cursor IDE Task agents on the **same account / machine / repo** work fine. So this appears specific to the **SDK Agent execution / streaming channel** , not our prompt or pack content.

* * *

## Environment

| Item | Value |
| --- | --- |
| OS | macOS 26.6 (Build 25G72), Darwin 25.6.0, arm64 (Apple M-series) |
| Host | GEIMACP4164J34CY |
| SDK | `cursor-sdk==1.0.31` (Python, latest at time of report) |
| Runtime | Python 3.14 in project venv `.venv-pilot` |
| Auth | `CURSOR_API_KEY` from Cursor Dashboard → Integrations (`crsr_…`) |
| Working dir | Local git repo (large monorepo; also reproduces with tiny prompt) |
| IDE | Cursor IDE Task sub-agents work on same machine/account |

* * *

## Expected vs actual

## Related public reports (similar shape)

- [SDK Agent.send streaming runs terminate RUNNING → ERROR with no diagnostic body (@cursor/sdk 1.0.22)](https://forum.cursor.com/t/sdk-agent-send-streaming-runs-terminate-running-error-with-no-diagnostic-body-cursor-sdk-1-0-22/167419)

Our status message string is specifically **`[unknown] [internal] Protocol error`** (not `stream_unavailable` / `ConnectError: [unavailable]`), but the RUNNING→ERROR-with-empty-body pattern matches.

* * *

## Asks for Cursor

1. Look up the run IDs above on your side — did the agent stream ever attach / complete?
2. Is this a known Agent API / transport regression for Integrations keys on 2026-09-17?
3. For Python `cursor-sdk`, is there an equivalent to JS `useHttp1ForAgent` / forced HTTP/1.1 we should set?
4. Any account/team entitlement required beyond `models.list` for Agent streaming?

* * *

## Workaround in use

Continue work via Cursor IDE Task sub-agents (`model=inherit` / Auto→Cost). Headless SDK queue is blocked until Protocol error clears.

### Steps to Reproduce

## Reproduction (minimal)

```python
import os
from pathlib import Path
from cursor_sdk import Agent, LocalAgentOptions, ModelSelection, ModelParameterValue

key = os.environ["CURSOR_API_KEY"]
model = ModelSelection(
    id="auto-smart",
    params=[ModelParameterValue(id="optimize_for", value="cost")],
)

with Agent.create(
    model=model,
    api_key=key,
    local=LocalAgentOptions(cwd=str(Path(".").resolve())),
) as agent:
    run = agent.send("Reply with exactly: OK")
    for msg in run.messages():
        if getattr(msg, "type", None) == "status":
            print(getattr(msg, "status", None), getattr(msg, "message", None))
    result = run.wait()
    print("final", result.status, result.id, repr(str(getattr(result, "result", "") or "")[:200]))

```

**Observed:**  
`RUNNING` → ~16s → `ERROR: [unknown] [internal] Protocol error` → empty result.

* * *

## What works vs what fails

| Check | Result |
| --- | --- |
| `Cursor.models.list(api_key=...)` | OK — includes `auto-smart` with optimize\_for `cost` / `balanced` / `intelligence` |
| Preflight / API key validation | OK |
| Cursor IDE Task agents (same account, same repo) | OK — used as workaround |
| SDK `auto-smart(cost)` local | **FAIL** Protocol error |
| SDK `auto-smart(balanced)` local | **FAIL** Protocol error |
| SDK `composer-2.5` local | **FAIL** Protocol error |
| SDK Cloud agent (`CloudAgentOptions`) | **FAIL** Protocol error |
| VPN on vs off | **FAIL** both ways |
| Prompt size / pack content | Ruled out — fails on `"Reply with exactly: OK"` |
| Multi-agent role switching (extractor→critic→fix) | Ruled out — fails on **first** agent only |

Note: Python `LocalAgentOptions` in 1.0.31 does **not** expose `useHttp1ForAgent` / `Cursor.configure({ local: ... })` (those appear in JS SDK forum guidance). We cannot apply that workaround from the Python client as documented for Node.

* * *

## Production failure context (same error class)

We run a sequential headless pilot queue (`Agent.create` → one role → next). Failures match the smoke test:

| When (UTC) | Role | Run ID | Context |
| --- | --- | --- | --- |
| 2026-09-17T23:08:19Z | critic | `run-63f55017-efcb-41f2-8623-79137e70c7a8` | #59 (VPN) |
| 2026-09-17T23:10:09Z | critic | `run-ef58aa8e-2007-4ce3-9585-6e863cbebcb2` | #59 (retry) |
| 2026-09-17T23:35:16Z | critic | `run-808eb2b8-c331-416c-8ecc-b7bc15c9ba67` | #59 (retry) |
| 2026-09-17T23:41:53Z | extractor | `run-43bde6ef-4860-480e-a747-09117acc5864` | #60 after pack init |
| 2026-09-18T00:00:01Z | extractor | `run-236b3d42-192a-45cb-9983-75c2f2f82ad8` | #61 after pack init |

Example transcript lines:

```plaintext
[extractor] status RUNNING:
[extractor] status ERROR: [unknown] [internal] Protocol error
[extractor] === agent run end status=error result= ===

```

Also captured agent IDs for the latest SIU failure:  
`agent-fc9e6b62-44b5-49ae-914f-293dc6f6d686` / `run-236b3d42-192a-45cb-9983-75c2f2f82ad8`.

Smoke-test run ID from earlier triage: `run-1c1c296c-aeca-496f-b911-25b5f0a122c1`.

* * *

### Expected Behavior

**Expected:** Agent stream progresses (assistant / tools / usage), then terminal success or an actionable typed error (auth, rate limit, quota, network).

**Actual:** Opaque `[unknown] [internal] Protocol error` after RUNNING; no diagnostic body; IDE path on same account still works.

* * *

### Operating System

MacOS

### Version Information

cursor-sdk 1.0.31

### For AI issues: which model did you use?

pl. refer to above

### Does this stop you from using Cursor

No - Cursor works, but with this issue

---

<div class="post-metadata">

**Author:** ![mohitjain](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/mohitjain/32/100229_2.png) [@mohitjain](https://forum.cursor.com/u/mohitjain)\
**Post date:** [September 18, 2026, 8:13am UTC](https://forum.cursor.com/t/cursor-sdk-failing-unknown-internal-protocol-error-15-20s/172083/7 "2026-09-18T08:13:27Z")

</div>

Hey @Krishna_Ponnamineni,

We checked every run and agent ID you listed - none reached our servers, and there are no errors on your account. Short version: local SDK agent runs stream over HTTP/2, while your `models.list` calls and the IDE use HTTP/1.1. Something on your network path (typically TLS inspection or a security proxy) is breaking the HTTP/2 stream, which surfaces as the `Protocol error`. It’s not your prompt, models, or account.

Fastest fix, in your control: have your network team exempt Cursor’s API domains (`*.cursor.sh`, especially `api2.cursor.sh`) from TLS/SSL inspection, or allow HTTP/2 bidirectional streaming to them. You can pinpoint it with the “Test HTTP/2 bidirectional streaming” curl here: [Network Configuration | Cursor Docs](https://cursor.com/docs/enterprise/network-configuration)

Quick answers:

1. Streams never attached, so there’s nothing more to pull for those IDs.
2. No Sep 17 regression - local SDK runs haven’t succeeded from this network before either.
3. Correct - the Python SDK has no HTTP/1.1 option like the JS `useHttp1ForAgent` yet. We’ve let the team know.
4. No extra entitlement needed beyond `models.list`.

The IDE Task agents are a solid stopgap in the meantime. We’re also looking into turning on HTTP/1.1 automatically for your org (no code change on your side) - I’ll follow up here once that’s in.

If the cloud agent (`CloudAgentOptions`) fails again, send the agent ID (starts with `bc-`) and timestamp; cloud uses a different path, so we’ll look at that one separately.

---

<div class="post-metadata">

**Author:** ![Krishna\_Ponnamineni](https://avatars.discourse-cdn.com/v4/letter/k/eb9ed0/32.png) [@Krishna\_Ponnamineni](https://forum.cursor.com/u/Krishna_Ponnamineni)\
**Post date:** [September 18, 2026, 8:56pm UTC](https://forum.cursor.com/t/cursor-sdk-failing-unknown-internal-protocol-error-15-20s/172083/8 "2026-09-18T20:56:01Z")

</div>

thank you for the feedback. I’ll switch to IDE task agents