dsh-llm-longcat
用于 DeepSeek Harness LLM 接口的 LongCat (LongCat-2.0) 适配器 — 100 万上下文、思考模式、工具调用
插件会安装到这里；不确定时保持 web。
npx -y @deepseek-ai/dsh plugin --profile web add github:ffyuuu/dsh-llm-longcat#a13729f78a26156b4f1d701dc9bd2ee09adbe5b7Llm Longcat 以 dsh-llm-longcat 发布，当前版本为 0.1.0。Plugin Hub 会校验它的 manifest，并保存精确安装来源，便于复现安装结果。
LongCat adapter for the DeepSeek Harness LLM seam.
Adds LongCat-2.0 as a model provider: 1M context, thinking mode, tool calling.
reasoning_content field and translates it into harness ReasoningBlocksarguments kept a raw JSON string end to endreasoning_content on tool-call turns, as thinking-mode passback requiresusage-before-finish ordering the harness relies onctx.credentials or the environment; no secret in any config file| Model | Context | Max output | Notes |
|---|---|---|---|
LongCat-2.0 | 1,048,576 | 131,072 | text-only; thinking + tool calling |
Facts from GET /openai/v1/models/LongCat-2.0, the only documented endpoint that
reports supported_parameters. Tool calling is not mentioned on the
chat-completions doc page and is only visible there.
dsh plugin --profile default add github:ffyuuu/dsh-llm-longcat
export LONGCAT_API_KEY=... # create one at https://longcat.chat/platform/api_keys
Installing a bundle lets the package's install scripts run on your machine, outside the sandbox the agent runs under. Pin a commit so a later push cannot change what executes:
dsh plugin --profile default add github:ffyuuu/dsh-llm-longcat#3dcb3b1b5870ba52baab053453bdbb28826e5f13
Then pick LongCat-2.0 in the model selector. The key may also be stored through the Web UI's Models page instead of the environment.
dsh itself will not installAt the time of writing, installing the harness can fail before any plugin is
reached, with either ETARGET … dsh-typert-protocol@^0.1.0-rc.8 or an npm
heap exhaustion. That is an upstream packaging state, not this plugin:
@deepseek-ai/dsh published 0.1.0-rc.8 while several packages it depends on
stopped at 0.1.0-rc.7, and because the manifests use caret ranges,
^0.1.0-rc.7 still resolves up into the missing rc.8. npm then backtracks
over an unsatisfiable graph until it runs out of memory.
Pinning every @deepseek-ai/* package to an exact 0.1.0-rc.7 through npm
overrides avoids the drift. Nothing in this plugin needs changing either
way — it declares >=0.1.0-rc.7 and works against whichever of those the
host ends up with.
- id: llm-longcat
name: dsh-llm-longcat
config:
apiKeyEnv: LONGCAT_API_KEY # default; resolved per request, never a literal key
baseURL: https://api.longcat.chat/openai/v1 # optional; $LONGCAT_BASE_URL then the public API
thinking: enabled # optional deployment policy; `disabled` locks every request to off
reasoningEffort: high # optional; off | high — LongCat's switch is binary
maxTokens: 131072 # optional per-request output cap
defaultContextWindow: 1048576
streamIdleTimeoutMs: 300000 # optional; five-minute default
retryPolicy: # optional; omission uses bounded normal defaults
mode: normal
maxRetries: 3
models:
- id: LongCat-2.0
contextWindow: 1048576
A llm-longcat: section in $DSH_HOME/settings.yaml overrides any field
without a restart: base URL, catalog, request defaults, and idle budget all
take effect on the next request, while an in-flight stream keeps the facts it
started with.
LongCat controls thinking with thinking: {type: enabled|disabled} and does
not accept OpenAI's top-level reasoning_effort — its
supported_parameters lists the former and omits the latter. There is
therefore no low/medium/high gradient to map, and this adapter offers exactly
two levels rather than advertising controls that would collapse onto the same
two request bodies:
| Selected effort | Wire body |
|---|---|
high ("Thinking") | {"thinking": {"type": "enabled"}} |
off | {"thinking": {"type": "disabled"}} |
| (none named) | resolves from config; still explicit |
off serializes an explicit disabled rather than omitting the field —
omitting it would hand the decision to LongCat's server-side default, which is
not what selecting Off should mean. Requesting low, medium, or max fails
with UNSUPPORTED_REASONING_EFFORT before any network I/O.
id and name as explicit null. LongCat sends
them on the opening delta and then null (not omitted) on every
continuation, so a naive !== undefined guard blanks the assembled call's
name. Verified on live traffic; pinned by a regression test.stream_options.include_usage always on. Usage may
arrive attached to the finish chunk or as a trailing usage-only chunk; both
are deferred to [DONE] so usage always precedes finish.reasoning_content is serialized back into history; on tool-call-free turns
it is dropped (ignored anyway — saves tokens).content is always a string, never null: the message is durable
session history, and a null there would make later turns replay a body the
endpoint can reject.prompt_tokens_details.cached_tokens maps to
cacheReadTokens and is subtracted out of inputTokens to keep the
harness's disjoint-count convention.Non-2xx responses throw LlmError with stable codes. LongCat documents a
dedicated 402 for exhausted token quota and puts insufficient_quota on
403, where most OpenAI-compatible providers use 429 — both are classified
as QUOTA before the auth and rate-limit buckets, so a depleted balance is
never reported as a bad key or retried as a transient rate limit.
| Condition | Code |
|---|---|
| 402, or quota detail at any status | QUOTA_EXCEEDED |
| 401 / 403 | AUTH |
| 429 | RATE_LIMIT |
| 400 with context-overflow detail | CONTEXT_WINDOW_EXCEEDED |
| other 400 | INVALID_REQUEST |
| 5xx | SERVER |
no [DONE] / bad JSON | STREAM_CLOSED / MALFORMED_RESPONSE |
A completed stream that opened no content blocks becomes a finish error with
EMPTY_RESPONSE, which the shipped retry policy treats as retryable.
npm run typecheck # against the published @deepseek-ai/dsh-llm types
npm test # 30 unit tests over serialize + translate
npm run build # emits lib/ and lib/types/
npm run test:e2e # real API, needs LONGCAT_API_KEY, spends a few hundred tokens
test:e2e drives the built adapter's own serialize → SSE → translate pipeline
against api.longcat.chat, so it verifies what the plugin actually sends
rather than a hand-written approximation. It is what caught the null-name
delta bug.
modality: text->text, so image
content is refused before sending, naming the model.stop is absent from supported_parameters; passing
one fails with UNSUPPORTED_OPTION rather than silently running past it.MIT