[Rrainich]

What this fixes

DeepSeek often answers HTTP 200 with a business-level error (code=40003 INVALID_TOKEN, "messages too frequent", a killed web session) rather than a 4xx. The bridge indexed straight into data.biz_data on those responses, which raised TypeError: 'NoneType' object is not subscriptable inside the route. No status was attached, so the token-rotation path never fired — the dead token stayed pinned in the pool and every later turn failed the same way in ~330 ms until it was removed by hand.

Changes

functions.py

• UpstreamBizError(UpstreamError) + _biz_data() — one helper for data.biz_data. Auth-shaped business messages (login|auth|token|session|expire) map to 401 so the token is quarantined/rotated; everything else maps to 429 so it cools down and rotates. Because the status rides on UpstreamError, _upstream_http_code() and the existing 401/403/429 path in handle_chat pick it up unchanged.
• Replaced every raw ["biz_data"] index: create_challange_pow, create_new_chat, upload_file (×2), get_file_content (×2).

app.py

• _setup_failure_outcome / _setup_retry_kwargs — create_new_chat / create_challange_pow run before the try block that owns rotation, so an auth failure there escaped handle_chat as a bare 500 and the dead token was never marked. These now classify the failure, mark the token, and re-enter handle_chat with the failed token excluded — outside the per-signature asyncio.Lock, which is not reentrant (recursing inside it hung the request).
• POST /v1/messages/count_tokens (plus the /messages/count_tokens alias) — the route did not exist, so every Claude Code context probe was a 404.
• _with_deadline — the shared ClientSession uses ClientTimeout(total=None, connect=15, sock_read=600), and sock_read resets on every keepalive byte, so a slowly-dripping upstream never tripped anything (the access log has 19.7-minute /v1/messages entries). This puts a configurable ceiling (DEEPSEEKER_REQUEST_DEADLINE, default 600 s) on upstream work and raises UpstreamError(504). Scope note: it is applied inside the SSE generators because they outlive handle_chat's return. It covers "upstream goes silent" and cancels the upstream generator; a slow client still surfaces as CancelledError rather than a 504, so it is not an end-to-end hard ceiling.
• _is_truncated_stream_error — upstream hang-ups mid-stream (aiohttp.ClientPayloadError / TransferEncodingError) logged a full traceback and left the client on a dead socket. Downgraded to a warning; the terminal error frame still goes out.

Tests

• tests/test_upstream_hardening.py — 15 cases: count_tokens auth + flattening, _biz_data status mapping, deadline pass-through / 504 on hang / no deadlock on early abort, truncation detection.
• tests/test_setup_retry.py — 2 cases: a 401 on create_new_chat rotates and succeeds instead of 500-ing, and the retry does not deadlock on the signature lock.

274 passed in 16.08s

Behaviour change to be aware of

A flagged/invalid token now produces an honest 401/429 and rotation instead of a wall of 500s. That is recovery, not repair: if every token belongs to the same flagged account, rotation has nowhere to go and the client sees a 429. That is the correct signal — the previous 500 loop was hiding it.

Not covered

• The rotation path itself has no live test against a genuinely flagged upstream account; it is covered at the unit level only.
• hardening-backup.patch in my working tree is a local rollback artifact and is not part of this PR.

---

Note: this supersedes #92, which was closed for an unrelated reason (a commit in that branch carried an unintended author identity; the branch was rewritten and re-pushed). The (PR #92) citations inside the CHANGELOG entries are from that history and refer to this same change set.

[Rrainich]

Credit where it's due: @1515asd diagnosed the parent_message_id: 0 rejection (biz_code 26, delivered as application/json on the SSE endpoint) independently, against a live incident, and published it with a fix in 1515asd/rh-deeperseeker (ac9a90f).

This PR carries the same two changes, implemented separately:

• First turn omits parent_message_id. Previously the code coerced 0 → null and always sent the key. It now omits the key when there's no parent, and only sends it on follow-up turns — matching the finding that omission is the only shape accepted across edges.
• Non-SSE tail diagnostics. The SSE parser keeps non-data: body lines and the response content-type, logs both when a stream ends without output, and promotes a JSON error envelope in that tail into the raised UpstreamError. Combined with this branch's UpstreamBizError mapping, that failure now carries a status instead of pinning the token.

Tests: tests/test_parent_message_id.py, 4 cases. Full suite 278 passed.

(This was originally opened as #92; that PR is closed and this one supersedes it with an identical diff.)

[AmanCode22]

Kindly fix merge conflicts and ruff check

[AmanCode22]

@Rrainich kindly fix ruff check and conflics of merge