baseten

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Baseten Product Overview

Baseten产品概述

Production AI inference platform - serve and scale open-source, custom, and fine-tuned models with the fastest runtimes, cross-cloud HA, and seamless developer workflows.

Dedicated Inference - deploy any model, performance-optimized + horizontally scaled. Authored as auto-wrapped Truss server, custom Docker server, or compound/orchestrated deployment via Chains.
Model APIs - pre-optimized hosted APIs for popular models. Path to graduate to dedicated.
Training - two paths: Truss Train (BYO container, any framework, full hardware control) and Loops (Tinker-compatible managed SDK for SFT + async RL; paired trainer + sampling server, live weight transfers, one-click checkpoint deploy). Multi-node, 1T+ params, 10TB+ datasets, H100/H200/B200. Remote access: SSH and VS Code/Cursor tunnels into containers.
Frontier Gateway - operate your own foundation model B2C.

生产级AI推理平台——凭借最快的运行时、跨云高可用（HA）以及流畅的开发者工作流，部署并扩展开源、自定义和微调模型。

专属推理 - 部署任意模型，性能优化+水平扩展。可通过自动封装的Truss服务器、自定义Docker服务器，或是借助Chains实现复合/编排式部署。
模型API - 针对热门模型预优化的托管API。可升级至专属部署。
训练 - 两种路径：Truss Train（自带容器，支持任意框架，完全硬件控制）和Loops（兼容Tinker的托管SDK，用于SFT+异步RL；配对训练器+采样服务器，实时权重传输，一键部署检查点）。支持多节点、1T+参数、10TB+数据集，适配H100/H200/B200硬件。远程访问：通过SSH和VS Code/Cursor隧道连接容器。
前沿网关（Frontier Gateway） - 运营您自己的大模型B2C业务。

Agent DX Toolkit

Agent DX工具包

Component	Provides	Install
`baseten` MCP	Interact with backend (~REST API, CRUD): models, deployments, training, environments, secrets, chains. API-key auth.	`npx add-mcp https://api.baseten.co/mcp -g -y --header "Authorization: Bearer ${BASETEN_MCP_KEY}"`
`baseten_docs` MCP	Semantic search + filesystem of `docs.baseten.co` . No auth.	`npx add-mcp https://docs.baseten.co/mcp -n "baseten_docs" -g -y`
`truss` CLI	Needed for model/chain push from local code, watch (= live patch). Needs `truss login` once.	`uv tool install truss` (or `pip install truss --upgrade` ; respect user package manager: uv, poetry...)
`llms.txt`	`baseten.co/llms.txt` (product + blog), `docs.baseten.co/llms.txt` (docs).	reachable via HTTP
This skill	`SKILL.md` + `references/*.md` loaded on demand.	`npx skills add basetenlabs/baseten-skills -g -y`

组件	功能	安装方式
`baseten` MCP	与后端交互（类REST API，CRUD操作）：模型、部署、训练、环境、密钥、Chains。API密钥认证。	`npx add-mcp https://api.baseten.co/mcp -g -y --header "Authorization: Bearer ${BASETEN_MCP_KEY}"`
`baseten_docs` MCP	`docs.baseten.co` 的语义搜索+文件系统。无需认证。	`npx add-mcp https://docs.baseten.co/mcp -n "baseten_docs" -g -y`
`truss` CLI	用于从本地代码推送模型/Chain、实时补丁（watch）。需执行一次 `truss login` 。	`uv tool install truss` （或 `pip install truss --upgrade` ；尊重用户包管理器：uv、poetry等）
`llms.txt`	`baseten.co/llms.txt` （产品+博客）、 `docs.baseten.co/llms.txt` （文档）。	可通过HTTP访问
本技能	按需加载 `SKILL.md` + `references/*.md` 。	`npx skills add basetenlabs/baseten-skills -g -y`

Setup

安装设置

Any subset works, full install recommended.
Suggest additional installs when the current task benefits from or requires them; help user with installation, but elicit preferences first.
Ensure
```
BASETEN_MCP_KEY
```
is provided when installing Baseten MCP (user can create key at
```
app.baseten.co/settings/api_keys
```
). Caveat: an MCP instance binds to one org/workspace at install time; switching the bound workspace later is not supported. To work with multiple workspaces, install additional MCP instances under different names with different keys (see last bullet of this section).
Truss CLI only needed for making deployments (check
```
truss --version
```
); prior login (multi-workspace users must provide
```
--remote <name>
```
). Explore with
```
truss [subcommand] --help
```
.
Docs MCP missing → grep / fetch
```
llms.txt
```
.
Backend MCP is API-key-only (currently); OAuth-only harnesses can still use the other components.
If backend MCP server is needed for different orgs/workspaces, add multiple MCP instances with different names/keys or use env-var expansion in the agent's config file and set the env var to the respective workspace's key.

可选择安装任意子集，推荐完整安装。
当当前任务需要或受益于额外组件时，建议用户安装；协助用户安装，但需先询问用户偏好。
安装Baseten MCP时需提供
```
BASETEN_MCP_KEY
```
（用户可在
```
app.baseten.co/settings/api_keys
```
创建密钥）。注意：MCP实例在安装时绑定到一个组织/工作区；后续不支持切换绑定的工作区。如需在多个工作区操作，请使用不同密钥安装多个同名不同配置的MCP实例（见本节最后一点）。
Truss CLI仅在创建部署时需要（检查
```
truss --version
```
）；首次登录（多工作区用户需指定
```
--remote <name>
```
）。可通过
```
truss [子命令] --help
```
查看详情。
若缺少Docs MCP → 搜索/获取
```
llms.txt
```
。
后端MCP目前仅支持API密钥认证；仅支持OAuth的工具仍可使用其他组件。
若需为不同组织/工作区使用后端MCP服务器，可添加多个不同名称/密钥的MCP实例，或在代理配置文件中使用环境变量扩展，并将环境变量设置为对应工作区的密钥。

Pick your authoring surface (for creating deployments)

选择部署创作方式

First-pass decision. Many real workloads blend rows — treat this as a starting point, not a rule. When unsure, sketch the IO shape and per-step hardware needs before picking.

Surfaces are stacked by opinion-strength, not just author convenience: engines (TRT-LLM, BEI, BIS-LLM) ship performance-tuned for one architecture and are the fastest path when they fit; custom Docker servers wrap mature inference servers (vLLM, SGLang, TGI, Triton, NIM); Python Truss is the escape hatch for arbitrary code in the request path; Chains add typed inter-step transport with built-in rate limiting, connection management, structured error propagation, and binary IO — features you'd otherwise rebuild around N raw Trusses. Python Truss and Chains share live-patch iteration (

truss watch

); all flavors support per-replica autoscaling, scale-to-zero, and environments / promotions.

You want…	Flavor	Specialization	When NOT to pick
Hosted LLM, no deploy step	Model APIs	`model-apis.md`	model not in catalog; need custom hardware, requirements, stability...
LLM on an off-the-shelf server (vLLM / SGLang / TGI / Triton / NIM)	Custom Docker server	`truss-custom-servers.md`	the server doesn't exist or you need Python in the request path
LLM/embedding via a Baseten engine (TRT-LLM / BEI / BIS-LLM); minimal config, no Python	Engine-only `config.yaml`	`truss-config.md` (engines section)	architecture not covered by an engine; you need custom logic
Custom Python in the request path (pre/post, custom arch, weird IO)	Python-class Truss ( `model.py` )	`truss-model-py.md` + `truss-config.md`	an engine or off-the-shelf server fits — pick that, it's faster to ship
Multi-step pipeline with heterogeneous hardware / per-step scaling (RAG, ASR→LLM→TTS, fan-out, chunking)	Chains	`truss-chains.md`	one-stage or homogeneous — a single Truss is simpler

Orthogonal operational surfaces (independent of which flavor above):

Iterate / patch a deployment →
```
model-dev-loop.md
```
Promote, environments, autoscaling →
```
deployment-lifecycle.md
```
Gradual traffic shift to new deployment (pause/resume/cancel) → rolling deployments:
```
deployment/rolling-deployments.mdx
```
Call a deployment →
```
inference-api.md
```
(custom) or
```
model-apis.md
```
(hosted)
High-throughput batch calls (embeddings, reranking) → Performance client (Rust, 1200+ req/s):
```
inference/performance-client.mdx
```
Programmatic control plane →
```
management-api.md
```

Real-world nuances the table can't capture:

Hybrids exist. A
```
model.py
```
can wrap an engine for pre/post-processing; a Chain entrypoint can be a Python class while internal Chainlets use engines.
Chain websockets are entrypoint-only. Intra-chainlet calls only stream output, but bi-di usually not needed on those edges.
Engine performance vs flexibility. TRT-LLM is the fastest path for many LLMs but its config surface is opaque. Worth the trade only when latency/throughput is a real constraint.

初步决策。许多实际工作负载会混合多种方式——将此视为起点，而非规则。不确定时，先勾勒IO形态和每一步的硬件需求再做选择。

各种方式按推荐优先级排序，而非仅考虑作者便利性：引擎（TRT-LLM、BEI、BIS-LLM）针对特定架构进行性能调优，适配时是最快路径；自定义Docker服务器封装成熟的推理服务器（vLLM、SGLang、TGI、Triton、NIM）；Python Truss是请求路径中任意代码的兜底方案；Chains添加了类型化的步骤间传输，内置限流、连接管理、结构化错误传播和二进制IO——这些功能若基于原生Truss实现则需自行开发。Python Truss和Chains支持实时补丁迭代（

truss watch

）；所有方式均支持每副本自动扩缩容、缩容至零，以及环境/版本升级。

您需要…	方式	参考文档	不适用场景
托管LLM，无需部署步骤	模型API	`model-apis.md`	模型不在目录中；需要自定义硬件、依赖、稳定性等...
在现成服务器上部署LLM（vLLM / SGLang / TGI / Triton / NIM）	自定义Docker服务器	`truss-custom-servers.md`	服务器不存在，或您需要在请求路径中加入Python代码
通过Baseten引擎（TRT-LLM / BEI / BIS-LLM）部署LLM/嵌入模型；配置极简，无需Python	仅引擎的 `config.yaml`	`truss-config.md` （引擎章节）	架构未被引擎覆盖；需要自定义逻辑
请求路径中加入自定义Python代码（预处理/后处理、自定义架构、特殊IO）	Python类Truss（ `model.py` ）	`truss-model-py.md` + `truss-config.md`	引擎或现成服务器可满足需求——优先选择，部署速度更快
多步骤流水线，异构硬件/按步骤扩缩容（RAG、ASR→LLM→TTS、扇出、分片）	Chains	`truss-chains.md`	单阶段或同构场景——单个Truss更简单

独立于上述方式的运维层面：

迭代/补丁部署 →
```
model-dev-loop.md
```
版本升级、环境管理、自动扩缩容 →
```
deployment-lifecycle.md
```
逐步将流量切换到新部署（暂停/恢复/取消）→ 滚动部署：
```
deployment/rolling-deployments.mdx
```
调用部署 →
```
inference-api.md
```
（自定义部署）或
```
model-apis.md
```
（托管部署）
高吞吐量批量调用（嵌入、重排序）→ 性能客户端（Rust，1200+请求/秒）：
```
inference/performance-client.mdx
```
程序化控制平面 →
```
management-api.md
```

表格未涵盖的实际场景细节：

混合方式存在。
```
model.py
```
可封装引擎以实现预处理/后处理；Chain入口可以是Python类，而内部Chainlet使用引擎。
Chain WebSocket仅支持入口。Chainlet内部调用仅支持输出流，通常这些边缘不需要双向流。
引擎性能vs灵活性。TRT-LLM是许多LLM的最快路径，但其配置界面不够透明。仅当延迟/吞吐量是实际约束时，才值得权衡。

Routing

路由指引

Skill References (

ls references/

in skill dir, complementary to hosted docs). Be generous to read any of the included reference files as soon as the user touches on that topic.

references/truss-cli.md

truss push

watch

/ iterate. Most-used. Deep dive:

references/truss-config.md

```
references/truss-model-py.md
```
: Python-class flavor (custom pre/post, non-engine architectures).
```
references/truss-custom-servers.md
```
:
```
docker_server
```
flavor (vLLM / TGI / SGLang / Triton; most common modern-LLM path).
```
references/truss-chains.md
```
: multi-step pipelines (RAG, ASR→LLM→TTS, chunked audio/video) with per-step HW + autoscaling.
```
references/model-apis.md
```
: shared pre-hosted endpoints (DeepSeek, GLM, Kimi, ...). Fastest when one fits.
```
references/inference-api.md
```
: calling custom deployments. connection reuse (
```
requests
```
/
```
httpx
```
/ OpenAI SDK), async / streaming / wake / OpenAI-compat sync routes.
```
references/management-api.md
```
: programmatic control plane (models, deployments, envs, secrets). What
```
truss
```
CLI uses under the hood.
```
references/deployment-lifecycle.md
```
: Model / Deployment / Environment semantics + promotion + autoscaling.
```
references/model-dev-loop.md
```
: post-first-deploy iteration. rebuild / patch / hot-reload cost tiers, agent-vs-human watch loop.

技能参考文档（技能目录下的

ls references/

，补充托管文档）。当用户涉及相关主题时，请主动阅读对应的参考文件。

references/truss-cli.md

：

truss push

watch

/ 迭代。最常用。深入了解：

references/truss-config.md

。

```
references/truss-model-py.md
```
：Python类方式（自定义预处理/后处理，非引擎架构）。
```
references/truss-custom-servers.md
```
：
```
docker_server
```
方式（vLLM / TGI / SGLang / Triton；现代LLM最常用路径）。
```
references/truss-chains.md
```
：多步骤流水线（RAG、ASR→LLM→TTS、音频/视频分片），支持按步骤分配硬件+自动扩缩容。
```
references/model-apis.md
```
：共享预托管端点（DeepSeek、GLM、Kimi等）。符合需求时是最快路径。
```
references/inference-api.md
```
：调用自定义部署。连接复用（
```
requests
```
/
```
httpx
```
/ OpenAI SDK）、异步/流式/唤醒/兼容OpenAI的同步路由。
```
references/management-api.md
```
：程序化控制平面（模型、部署、环境、密钥）。
```
truss
```
CLI底层依赖的接口。
```
references/deployment-lifecycle.md
```
：模型/部署/环境语义 + 版本升级 + 自动扩缩容。
```
references/model-dev-loop.md
```
：首次部署后的迭代。重建/补丁/热重载的成本层级，代理vs人工watch循环。

Gotchas

注意事项

Don't speculate, query

不要猜测，要查询

For any perf/status/error claim, use the tools first — don't estimate or guess.

Timings → deployment / chainlet log tools (timestamped, includes
```
Pulling image
```
,
```
model_cache: Fetch took
```
,
```
Completed model.load() execution in N ms
```
, per-request markers).
Status → deployment-get tools before invoking.
Build/deploy failure → fetch logs immediately, don't hypothesize.

Fabricating numbers from training-data priors burns user trust; logs are the source of truth.

对于任何性能/状态/错误问题，先使用工具——不要估算或猜测。

计时信息 → 部署/chainlet日志工具（带时间戳，包含
```
Pulling image
```
、
```
model_cache: Fetch took
```
、
```
Completed model.load() execution in N ms
```
、每个请求的标记）。
状态信息 → 调用前先使用部署查询工具。
构建/部署失败 → 立即获取日志，不要假设原因。

根据训练数据臆造数据会损害用户信任；日志是唯一的事实来源。

Source heterogeneity & drift

来源异构性与偏差

Content lives across systems that don't overlap cleanly and drift independently. No single source is perfect/authoritative. For any non-trivial claim ("supported", perf numbers, recommended approach), triangulate across ≥2 sources. Surface contradictions to the user; don't paper over.

Source	Strength	Gap / quirk
`baseten_docs` MCP / `docs.baseten.co`	API specs, protocol details, knobs	Lags product; no perf numbers
`baseten.co/library/<id>` (marketing)	Flagship managed models, perf claims	Some entries are sales-gated, not self-serve
`baseten` MCP `list_library_models`	What's actually one-click API-deployable	Doesn't include every marketing-library entry; lacking tags
`baseten.co/blog/`	Concrete latency / cost / vs-competitor numbers, technical deep dives	Unstructured; not in docs MCP. Discover via `baseten.co/llms.txt`
`baseten.co/solutions/...`	High-level pitch	May describe flagship features that need a Baseten engagement
`truss-examples` GitHub repo	Working code patterns	Often outdated / broken / drifted. Consult with caution, last resort. Might need fixups before deploy works.

Library page exists but model absent from
```
list_library_models
```
→ likely managed/flagship; tell the user it may need to reach out to Baseten support.
Perf numbers found only in a blog → cite as blog claim.
Can't find something via docs MCP → fetch
```
baseten.co/llms.txt
```
or
```
docs.baseten.co/llms.txt
```
as index, then fetch the page directly. Last resort: web search.

内容分布在多个系统中，这些系统并非完全重叠，且各自独立更新。没有单一来源是完美/权威的。对于任何非琐碎的声明（“支持”、性能数据、推荐方案），至少从2个来源交叉验证。向用户指出矛盾之处；不要掩盖。

来源	优势	不足/特殊点
`baseten_docs` MCP / `docs.baseten.co`	API规范、协议细节、配置项	滞后于产品；无性能数据
`baseten.co/library/<id>` （营销页面）	旗舰托管模型、性能声明	部分内容需销售对接，无法自助使用
`baseten` MCP `list_library_models`	实际可一键API部署的模型	不包含所有营销库中的模型；缺少标签
`baseten.co/blog/`	具体延迟/成本/竞品对比数据、技术深度解析	非结构化；不在docs MCP中。可通过 `baseten.co/llms.txt` 发现
`baseten.co/solutions/...`	高层级方案介绍	可能描述的是需要Baseten对接的旗舰功能
`truss-examples` GitHub仓库	可用的代码模式	经常过时/损坏/偏离当前版本。谨慎参考，仅作为最后手段。部署前可能需要修复。

库页面存在但模型未在
```
list_library_models
```
中 → 可能是托管/旗舰模型；告知用户可能需要联系Baseten支持。
仅在博客中找到性能数据 → 注明是博客中的声明。
无法通过docs MCP找到内容 → 获取
```
baseten.co/llms.txt
```
或
```
docs.baseten.co/llms.txt
```
作为索引，然后直接获取页面。最后手段：网页搜索。

Non-obvious placements within

references/

references/

中不明显的内容位置

Engine-only deploys (TensorRT-LLM, BEI, BIS-LLM) →
```
truss-config.md
```
engines section (also owns
```
model_cache
```
, secrets, resources).
Authoring-flavor decision: single deployment → top of
```
truss-config.md
```
; multiple coordinated →
```
truss-chains.md
```
.
Training and Frontier Gateway: no skill reference. Use
```
baseten
```
MCP +
```
baseten_docs
```
MCP. For training path choice see
```
training/overview.mdx
```
; for Loops (managed SFT/RL SDK) see
```
loops/overview.mdx
```
; for SSH / VS Code tunnels into training containers see
```
training/ssh.mdx
```
and
```
training/remote-access.mdx
```
.

仅引擎部署（TensorRT-LLM、BEI、BIS-LLM）→
```
truss-config.md
```
引擎章节（同时包含
```
model_cache
```
、密钥、资源配置）。
创作方式决策：单部署 →
```
truss-config.md
```
顶部；多协同部署 →
```
truss-chains.md
```
。
训练和前沿网关：无技能参考文档。使用
```
baseten
```
MCP +
```
baseten_docs
```
MCP。训练路径选择见
```
training/overview.mdx
```
；Loops（托管SFT/RL SDK）见
```
loops/overview.mdx
```
；训练容器的SSH / VS Code隧道见
```
training/ssh.mdx
```
和
```
training/remote-access.mdx
```
。

Tool quirks

工具特性

```
baseten_docs
```
MCP search is semantic (embedding-based). Results are generally intent-aware, but can still miss on niche phrasing — try rephrased queries or fetch
```
docs.baseten.co/llms.txt
```
as an index if results are weak.
Full doc pages:
```
cat
```
/
```
head
```
on
```
.mdx
```
paths via
```
query_docs_filesystem_baseten
```
. Alternatively,
```
docs.baseten.co/<path>.md
```
also works and is faster for single-page reads.
Inference SSH (terminal in running container, requires org enablement) →
```
model-dev-loop.md
```
§ Inference SSH.
```
list_library_models
```
is baseten curated catalog (~tens of pre-optimized hosted models, mostly popular open-source LLMs / embeddings). Models are good starter models to play, but not for custom authoring, specific performance needs, finetuning and private HF models etc. No useful tags (modality etc.) — filter by
```
display_name
```
/
```
hf_repo_id
```
substring.
Blog content is not in the docs MCP. Fetch
```
baseten.co/llms.txt
```
and search for relevant posts.

```
baseten_docs
```
MCP搜索是语义化的（基于嵌入）。结果通常能匹配意图，但仍可能遗漏小众表述——若结果不佳，尝试重新表述查询或获取
```
docs.baseten.co/llms.txt
```
作为索引。
完整文档页面：通过
```
query_docs_filesystem_baseten
```
对
```
.mdx
```
路径执行
```
cat
```
/
```
head
```
操作。或者，
```
docs.baseten.co/<path>.md
```
也可用，单页面读取速度更快。
推理SSH（运行中容器的终端，需组织启用）→
```
model-dev-loop.md
```
中的“推理SSH”章节。
```
list_library_models
```
是Baseten curated目录（约数十个预优化的托管模型，多为热门开源LLM/嵌入模型）。这些模型适合入门试用，但不适用于自定义创作、特定性能需求、微调及私有HF模型等场景。无有用标签（如模态等）——可通过
```
display_name
```
/
```
hf_repo_id
```
子串过滤。
博客内容不在docs MCP中。获取
```
baseten.co/llms.txt
```
并搜索相关文章。

MCP introspection vs. invocation

MCP内省与调用

Backend MCP tools (

get_deployment

get_deployment_logs

get_deployment_config

list_*

, …) describe state. They are strongest when you need an exact id / schema / config value to act on, and weakest when the user's complaint is about runtime behavior. Resist letting MCP availability shrink the size of your investigation.

Look-up tasks (need an ID, current config, or schema before writing a patch): MCP get/list is the right starting point. Reading the live config beats inferring it.
Behavior tasks (broken / slow / wrong-shape responses, anything the user observed): MCP can supplement but rarely substitutes for exercising the endpoint yourself. A
```
curl
```
against the predict URL with a representative payload runs the same code path the user does. Absence of recent errors in MCP-fetched logs is not proof the bug is gone — log windows are bounded, and a problem the user reported five minutes ago may have already aged out of the default tail. For a specific historic incident you can't reproduce, log tools support custom time ranges within retention; use them.
Treat cheap introspection as budget freed up for deeper verification (re-curl after a fix, diff configs before and after a PATCH, push a candidate truss and probe it), not as license to stop earlier.

后端MCP工具（

get_deployment

、

get_deployment_logs

、

get_deployment_config

、

list_*

等）描述状态。当您需要确切的ID/ schema/配置值来执行操作时，它们最有用；当用户反馈的是运行时行为问题时，它们作用最弱。不要因为MCP可用就缩小调查范围。

查询任务（需要ID、当前配置或schema来编写补丁）：MCP的get/list是正确起点。读取实时配置优于推断。
行为任务（响应异常/缓慢/格式错误，用户观察到的任何问题）：MCP可作为补充，但很少能替代您自己调用端点测试。使用代表性payload对预测URL执行
```
curl
```
请求，走的是与用户相同的代码路径。MCP获取的日志中没有近期错误，不能证明问题已解决——日志窗口是有限的，用户5分钟前报告的问题可能已超出默认日志保留范围。对于无法重现的特定历史事件，日志工具支持在保留期内自定义时间范围；请使用该功能。
将低成本的内省视为可用于深度验证的预算（修复后重新curl，PATCH前后对比配置，推送候选truss并测试），而非提前结束调查的理由。

baseten

Original

Translation

Baseten Product Overview

Baseten产品概述

Agent DX Toolkit

Agent DX工具包

Setup

安装设置

Pick your authoring surface (for creating deployments)

选择部署创作方式

Routing

路由指引

Gotchas

注意事项

Don't speculate, query

不要猜测，要查询

Source heterogeneity & drift

来源异构性与偏差

Non-obvious placements within
`references/`

`references/`
中不明显的内容位置

Tool quirks

工具特性

MCP introspection vs. invocation

MCP内省与调用

baseten

Original

Translation

Baseten Product Overview

Baseten产品概述

Agent DX Toolkit

Agent DX工具包

Setup

安装设置

Pick your authoring surface (for creating deployments)

选择部署创作方式

Routing

路由指引

Gotchas

注意事项

Don't speculate, query

不要猜测，要查询

Source heterogeneity & drift

来源异构性与偏差

Non-obvious placements within references/

references/中不明显的内容位置

Tool quirks

工具特性

MCP introspection vs. invocation

MCP内省与调用

Non-obvious placements within
`references/`

`references/`
中不明显的内容位置