# node-newrelic - Doramagic AI Context Pack

> 定位：安装前体验与判断资产。它帮助宿主 AI 有一个好的开始，但不代表已经安装、执行或验证目标项目。

## 充分原则

- **充分原则，不是压缩原则**：AI Context Pack 应该充分到让宿主 AI 在开工前理解项目价值、能力边界、使用入口、风险和证据来源；它可以分层组织，但不以最短摘要为目标。
- **压缩策略**：只压缩噪声和重复内容，不压缩会影响判断和开工质量的上下文。

## 给宿主 AI 的使用方式

你正在读取 Doramagic 为 node-newrelic 编译的 AI Context Pack。请把它当作开工前上下文：帮助用户理解适合谁、能做什么、如何开始、哪些必须安装后验证、风险在哪里。不要声称你已经安装、运行或执行了目标项目。

## Claim 消费规则

- **事实来源**：Repo Evidence + Claim/Evidence Graph；Human Wiki 只提供显著性、术语和叙事结构。
- **事实最低状态**：`supported`
- `supported`：可以作为项目事实使用，但回答中必须引用 claim_id 和证据路径。
- `weak`：只能作为低置信度线索，必须要求用户继续核实。
- `inferred`：只能用于风险提示或待确认问题，不能包装成项目事实。
- `unverified`：不得作为事实使用，应明确说证据不足。
- `contradicted`：必须展示冲突来源，不得替用户强行选择一个版本。

## 它最适合谁

- **想在安装前理解开源项目价值和边界的用户**：当前证据主要来自项目文档。 证据：`README.md` Claim：`clm_0002` supported 0.86

## 它能做什么

- **命令行启动或安装流程**（需要安装后验证）：项目文档中存在可执行命令，真实使用需要在本地或宿主环境中运行这些命令。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0001` supported 0.86

## 怎么开始

- `curl http://localhost:3000/api/users/abc123?id=true` 证据：`NEWS.md` Claim：`clm_0003` unverified 0.25
- `npx --package=protobufjs-cli -c 'pbjs --keep-case v1.proto' > v1.json` 证据：`lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0004` supported 0.86
- `curl https://www.apache.org/dist/cassandra/KEYS | sudo apt-key add -` 证据：`bin/cassandra-setup.sh` Claim：`clm_0005` unverified 0.25
- `git clone git@source.datanerd.us:newrelic/cross_agent_tests.git test/lib/cross_agent_tests` 证据：`bin/update-cats.sh` Claim：`clm_0006` unverified 0.25

## 继续前判断卡

- **当前建议**：先做角色匹配试用
- **为什么**：这个项目更像角色库，核心风险是选错角色或把角色文案当执行能力；先用 Prompt Preview 试角色匹配，再决定是否沙盒导入。

### 30 秒判断

- **现在怎么做**：先做角色匹配试用
- **最小安全下一步**：先用 Prompt Preview 试角色匹配；满意后再隔离导入
- **先别相信**：角色质量和任务匹配不能直接相信。
- **继续会触碰**：角色选择偏差、命令执行、本地环境或项目文件

### 现在可以相信

- **适合人群线索：想在安装前理解开源项目价值和边界的用户**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`README.md` Claim：`clm_0002` supported 0.86
- **能力存在：命令行启动或安装流程**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0001` supported 0.86
- **存在 Quick Start / 安装命令线索**（supported）：可以相信项目文档出现过启动或安装入口；不要因此直接在主力环境运行。 证据：`lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0004` supported 0.86

### 现在还不能相信

- **角色质量和任务匹配不能直接相信。**（unverified）：角色库证明有很多角色，不证明每个角色都适合你的具体任务，也不证明角色能产生高质量结果。
- **不能把角色文案当成真实执行能力。**（unverified）：安装前只能判断角色描述和任务画像是否匹配，不能证明它能在宿主 AI 里完成任务。
- **真实输出质量不能在安装前相信。**（unverified）：Prompt Preview 只能展示引导方式，不能证明真实项目中的结果质量。
- **宿主 AI 版本兼容性不能在安装前相信。**（unverified）：Claude、Cursor、Codex、Gemini 等宿主加载规则和版本差异必须在真实环境验证。
- **不会污染现有宿主 AI 行为，不能直接相信。**（inferred）：Skill、plugin、AGENTS/CLAUDE/GEMINI 指令可能改变宿主 AI 的默认行为。
- **可安全回滚不能默认相信。**（unverified）：除非项目明确提供卸载和恢复说明，否则必须先在隔离环境验证。
- **真实安装后是否与用户当前宿主 AI 版本兼容？**（unverified）：兼容性只能通过实际宿主环境验证。
- **项目输出质量是否满足用户具体任务？**（unverified）：安装前预览只能展示流程和边界，不能替代真实评测。

### 继续会触碰什么

- **角色选择偏差**：用户对任务应该由哪个专家角色处理的判断。 原因：选错角色会让 AI 从错误专业视角回答，浪费时间或误导决策。
- **命令执行**：包管理器、网络下载、本地插件目录、项目配置或用户主目录。 原因：运行第一条命令就可能产生环境改动；必须先判断是否值得跑。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md`
- **本地环境或项目文件**：安装结果、插件缓存、项目配置或本地依赖目录。 原因：安装前无法证明写入范围和回滚方式，需要隔离验证。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md`
- **宿主 AI 上下文**：AI Context Pack、Prompt Preview、Skill 路由、风险规则和项目事实。 原因：导入上下文会影响宿主 AI 后续判断，必须避免把未验证项包装成事实。

### 最小安全下一步

- **先跑 Prompt Preview**：先用交互式试用验证任务画像和角色匹配，不要先导入整套角色库。（适用：任何项目都适用，尤其是输出质量未知时。）
- **只在隔离目录或测试账号试装**：避免安装命令污染主力宿主 AI、真实项目或用户主目录。（适用：存在命令执行、插件配置或本地写入线索时。）
- **安装后只验证一个最小任务**：先验证加载、兼容、输出质量和回滚，再决定是否深用。（适用：准备从试用进入真实工作流时。）

### 退出方式

- **保留安装前状态**：记录原始宿主配置和项目状态，后续才能判断是否可恢复。
- **保留原始角色选择记录**：如果输出偏题，可以回到任务画像阶段重新选择角色，而不是继续沿着错误角色推进。
- **记录安装命令和写入路径**：没有明确卸载说明时，至少要知道哪些目录或配置需要手动清理。
- **如果没有回滚路径，不进入主力环境**：不可回滚是继续前阻断项，不应靠信任或运气继续。

## 哪些只能预览

- 解释项目适合谁和能做什么
- 基于项目文档演示典型对话流程
- 帮助用户判断是否值得安装或继续研究

## 哪些必须安装后验证

- 真实安装 Skill、插件或 CLI
- 执行脚本、修改本地文件或访问外部服务
- 验证真实输出质量、性能和兼容性

## 边界与风险判断卡

- **把安装前预览误认为真实运行**：用户可能高估项目已经完成的配置、权限和兼容性验证。 处理方式：明确区分 prompt_preview_can_do 与 runtime_required。 Claim：`clm_0007` inferred 0.45
- **命令执行会修改本地环境**：安装命令可能写入用户主目录、宿主插件目录或项目配置。 处理方式：先在隔离环境或测试账号中运行。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0008` supported 0.86
- **待确认**：真实安装后是否与用户当前宿主 AI 版本兼容？。原因：兼容性只能通过实际宿主环境验证。
- **待确认**：项目输出质量是否满足用户具体任务？。原因：安装前预览只能展示流程和边界，不能替代真实评测。
- **待确认**：安装命令是否需要网络、权限或全局写入？。原因：这影响企业环境和个人环境的安装风险。

## 开工前工作上下文

### 加载顺序

- 先读取 how_to_use.host_ai_instruction，建立安装前判断资产的边界。
- 读取 claim_graph_summary，确认事实来自 Claim/Evidence Graph，而不是 Human Wiki 叙事。
- 再读取 intended_users、capabilities 和 quick_start_candidates，判断用户是否匹配。
- 需要执行具体任务时，优先查 role_skill_index，再查 evidence_index。
- 遇到真实安装、文件修改、网络访问、性能或兼容性问题时，转入 risk_card 和 boundaries.runtime_required。

### 任务路由

- **命令行启动或安装流程**：先说明这是安装后验证能力，再给出安装前检查清单。 边界：必须真实安装或运行后验证。 证据：`NEWS.md`, `bin/cassandra-setup.sh`, `bin/update-cats.sh`, `lib/grpc/endpoints/infinite-tracing/Readme.md` Claim：`clm_0001` supported 0.86

### 上下文规模

- 文件总数：1909
- 重要文件覆盖：40/1909
- 证据索引条目：80
- 角色 / Skill 条目：28

### 证据不足时的处理

- **missing_evidence**：说明证据不足，要求用户提供目标文件、README 段落或安装后验证记录；不要补全事实。
- **out_of_scope_request**：说明该任务超出当前 AI Context Pack 证据范围，并建议用户先查看 Human Manual 或真实安装后验证。
- **runtime_request**：给出安装前检查清单和命令来源，但不要替用户执行命令或声称已执行。
- **source_conflict**：同时展示冲突来源，标记为待核实，不要强行选择一个版本。

## Prompt Recipes

### 适配判断

- 目标：判断这个项目是否适合用户当前任务。
- 预期输出：适配结论、关键理由、证据引用、安装前可预览内容、必须安装后验证内容、下一步建议。

```text
请基于 node-newrelic 的 AI Context Pack，先问我 3 个必要问题，然后判断它是否适合我的任务。回答必须包含：适合谁、能做什么、不能做什么、是否值得安装、证据来自哪里。所有项目事实必须引用 evidence_refs、source_paths 或 claim_id。
```

### 安装前体验

- 目标：让用户在安装前感受核心工作流，同时避免把预览包装成真实能力或营销承诺。
- 预期输出：一段带边界标签的体验剧本、安装后验证清单和谨慎建议；不含真实运行承诺或强营销表述。

```text
请把 node-newrelic 当作安装前体验资产，而不是已安装工具或真实运行环境。

请严格输出四段：
1. 先问我 3 个必要问题。
2. 给出一段“体验剧本”：用 [安装前可预览]、[必须安装后验证]、[证据不足] 三种标签展示它可能如何引导工作流。
3. 给出安装后验证清单：列出哪些能力只有真实安装、真实宿主加载、真实项目运行后才能确认。
4. 给出谨慎建议：只能说“值得继续研究/试装”“先补充信息后再判断”或“不建议继续”，不得替项目背书。

硬性边界：
- 不要声称已经安装、运行、执行测试、修改文件或产生真实结果。
- 不要写“自动适配”“确保通过”“完美适配”“强烈建议安装”等承诺性表达。
- 如果描述安装后的工作方式，必须使用“如果安装成功且宿主正确加载 Skill，它可能会……”这种条件句。
- 体验剧本只能写成“示例台词/假设流程”：使用“可能会询问/可能会建议/可能会展示”，不要写“已写入、已生成、已通过、正在运行、正在生成”。
- Prompt Preview 不负责给安装命令；如用户准备试装，只能提示先阅读 Quick Start 和 Risk Card，并在隔离环境验证。
- 所有项目事实必须来自 supported claim、evidence_refs 或 source_paths；inferred/unverified 只能作风险或待确认项。

```

### 角色 / Skill 选择

- 目标：从项目里的角色或 Skill 中挑选最匹配的资产。
- 预期输出：候选角色或 Skill 列表，每项包含适用场景、证据路径、风险边界和是否需要安装后验证。

```text
请读取 role_skill_index，根据我的目标任务推荐 3-5 个最相关的角色或 Skill。每个推荐都要说明适用场景、可能输出、风险边界和 evidence_refs。
```

### 风险预检

- 目标：安装或引入前识别环境、权限、规则冲突和质量风险。
- 预期输出：环境、权限、依赖、许可、宿主冲突、质量风险和未知项的检查清单。

```text
请基于 risk_card、boundaries 和 quick_start_candidates，给我一份安装前风险预检清单。不要替我执行命令，只说明我应该检查什么、为什么检查、失败会有什么影响。
```

### 宿主 AI 开工指令

- 目标：把项目上下文转成一次对话开始前的宿主 AI 指令。
- 预期输出：一段边界明确、证据引用明确、适合复制给宿主 AI 的开工前指令。

```text
请基于 node-newrelic 的 AI Context Pack，生成一段我可以粘贴给宿主 AI 的开工前指令。这段指令必须遵守 not_runtime=true，不能声称项目已经安装、运行或产生真实结果。
```

## 角色 / Skill 索引

- 共索引 28 个角色 / Skill / 项目文档条目。

- **New Relic's Node.js agent**（project_doc）：New Relic's Node.js agent ! npm status badge 1 2 ! Server Smoke Tests 3 4 ! Node Agent CI 5 6 ! codecov 7 8 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`README.md`
- **Node Agent Cloud Tooling**（project_doc）：This repository contains various cloud based tools for the New Relic Node Agent e.g. Azure Site Extensions . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`cloud-tooling/README.md`
- **Readme**（project_doc）：This directory contains resources for Docker containers that are used in our versioned tests. A typical case would be to mount a directory as a volume. For example: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docker/Readme.md`
- **Azure Node Agent Site Extension**（project_doc）：This project creates an Azure site extension that automatically installs the New Relic Node Agent. This extension is designed for Node applications running on Azure Windows compute resources. The site extensions follow semantic versioning conventions https://semver.org/ . You can expect to find artifacts in Nuget https://www.nuget.org/ . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`cloud-tooling/azure-site-extension/Content/README.md`
- **Readme**（project_doc）：The .proto files in this directory are provided by New Relic internal specifications. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`lib/grpc/endpoints/infinite-tracing/Readme.md`
- **Sampling at New Relic**（project_doc）：The New Relic agent supports a robust sampling decision making interface. This is a work-in-progress feature. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`lib/samplers/README.md`
- **Subscriber-based Instrumentation**（project_doc）：As of v13.2.0, we have begun to refactor our traditional instrumentation Shim -based monkey-patching to instead subscribe to events emitted by Node's diagnostic channel TracingChannel https://nodejs.org/api/diagnostics channel.html class-tracingchannel . This is done through @apm-js-collab/tracing-hooks https://github.com/apm-js-collab/tracing-hooks , a helper for orchestrion-js https://github.com/apm-js-collab/orch… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`lib/subscribers/README.md`
- **Running Benchmarks**（project_doc）：The easiest way to run all benchmarks is by using the npm script: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/benchmark/Readme.md`
- **Readme**（project_doc）：The protobuf schemas in this directory are copied from https://github.com/open-telemetry/opentelemetry-proto/tree/68f9c63/opentelemetry/proto 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/integration/otel/schemas/Readme.md`
- **Cross Agent Tests**（project_doc）：None of these tests should contain customer data such as SQL strings. Please be careful when adding new tests from real world failures. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/README.md`
- **CAT Map test details**（project_doc）：The CAT map test cases in cat map.json are meant to be used to verify the attributes that agents collect and attach to analytics transaction events for the CAT map project. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/cat/README.md`
- **Datastore instance tests**（project_doc）：The datastore instance tests provide attributes similar to what an agent could expect to find regarding a database configuration and specifies the expected datastore instance metric https://source.datanerd.us/agents/agent-specs/blob/master/Datastore-Metrics-PORTED.md datastore-metric-namespace that should be generated. The table below lists types attributes and whether they will always be included or optionally incl… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/datastores/README.md`
- **Trace Context test details**（project_doc）：The Trace Context test cases in trace context.json are meant to be used to verify the creation and forwarding of W3C Trace Context headers within the agent and the attributes and metrics that get created during that process. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/distributed_tracing/README.md`
- **Readme**（project_doc）：These tests cover parsing of Docker container IDs on Linux hosts out of /proc/self/cgroup or /proc/ /cgroup more generally . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/docker_container_id/README.md`
- **Readme**（project_doc）：These tests cover parsing of Docker container IDs on Linux hosts out of /proc/self/mountinfo or /proc/ /mountinfo more generally . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/docker_container_id_v2/README.md`
- **event source info**（project_doc）：This test fixture is intended to verify that the the agent correctly detects event types in languages where type is dynamic and correctly harvests ARN values. The fixtures are structured as follows: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/lambda/README.md`
- **PostgreSQL explain plan obfuscation tests**（project_doc）：PostgreSQL explain plan obfuscation tests 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/postgres_explain_obfuscation/README.md`
- **Readme**（project_doc）：These tests are for determining the numbers of physical packages, physical cores, and logical processors from the data returned by /proc/cpuinfo on Linux hosts. Each text file in this directory is the output of /proc/cpuinfo on various machines. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/proc_cpuinfo/README.md`
- **Readme**（project_doc）：These tests are for determining the physical memory from the data returned by /proc/meminfo on Linux hosts. The total physical memory of the linux system is reported as part of the environment values. The key used by the Python agent is 'Total Physical Memory MB '. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/proc_meminfo/README.md`
- **Samplers**（project_doc）：With the introduction of Otel-style sampling algorithms and core tracing, we now have many configurable samplers that may be working simultaneously within one application. These tests describe how the samplers should be set up based on local config, and how they are expected to behave under traffic. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/samplers/README.md`
- **Readme**（project_doc）：These test cases cover obfuscation more properly, masking of literal values from SQL statements captured by agents. SQL statements may be captured and attached to transaction trace nodes, or to slow SQL traces. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/sql_obfuscation/README.md`
- **Synthetics Tests**（project_doc）：The Synthetics tests are designed to verify that the agent handles valid and invalid Synthetics requests. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/synthetics/README.md`
- **The Utilization Tests**（project_doc）：The Utilization tests ensure that the appropriate information is being gathered for pricing. It is centered around ensuring that the JSON generated by all agents is correct. Each JSON block is a test case, with potentially the following fields: - testname : The name of the test. - input total ram mib : The total ram number calculated by the agent. - input logical processors : The number of logical processors calcula… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/utilization/README.md`
- **Vendor Specific Utilization Tests**（project_doc）：The Utilization tests ensure that the appropriate information is being gathered for pricing for a particular cloud vendor. It is centered around ensuring that the JSON generated by all agents is correct. Each JSON block is a test case, with potentially the following fields: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/cross_agent_tests/utilization_vendor_specific/README.md`
- **Readme**（project_doc）：This module is a collection of helper functions to assert expectations in tests. Each file represents a single assertion function, and must export a named function. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/lib/custom-assertions/Readme.md`
- **Readme**（project_doc）：This folder contains a very basic Express server that is instrumented with the Node.js agent. It will run a test suite that just makes a http requests and verifies that it does not crash the server and properly responds. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/smoke/e2e/README.md`
- **Versioned Tests**（project_doc）：The subdirectories within this directory represent test suites that verify our agent compatibility across multiple versions of our supported modules. When these tests are run, through our versioned test runner tool, each suite is run multiple times utilizing different versions of the module that satisfy configured version constraints. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/versioned/Readme.md`
- **Readme**（project_doc）：As of 2024-10-30, the consensus on the team is that this test exists to ensure we are not breaking applications that utilize the continuation-local-storage module. We do not instrument that module, but we shouldn't be breaking it, either. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`test/versioned/cls/Readme.md`

## 证据索引

- 共索引 80 条证据。

- **New Relic's Node.js agent**（documentation）：New Relic's Node.js agent ! npm status badge 1 2 ! Server Smoke Tests 3 4 ! Node Agent CI 5 6 ! codecov 7 8 证据：`README.md`
- **Node Agent Cloud Tooling**（documentation）：This repository contains various cloud based tools for the New Relic Node Agent e.g. Azure Site Extensions . 证据：`cloud-tooling/README.md`
- **Readme**（documentation）：This directory contains resources for Docker containers that are used in our versioned tests. A typical case would be to mount a directory as a volume. For example: 证据：`docker/Readme.md`
- **Azure Node Agent Site Extension**（documentation）：This project creates an Azure site extension that automatically installs the New Relic Node Agent. This extension is designed for Node applications running on Azure Windows compute resources. The site extensions follow semantic versioning conventions https://semver.org/ . You can expect to find artifacts in Nuget https://www.nuget.org/ . 证据：`cloud-tooling/azure-site-extension/Content/README.md`
- **Readme**（documentation）：The .proto files in this directory are provided by New Relic internal specifications. 证据：`lib/grpc/endpoints/infinite-tracing/Readme.md`
- **Sampling at New Relic**（documentation）：The New Relic agent supports a robust sampling decision making interface. This is a work-in-progress feature. 证据：`lib/samplers/README.md`
- **Subscriber-based Instrumentation**（documentation）：As of v13.2.0, we have begun to refactor our traditional instrumentation Shim -based monkey-patching to instead subscribe to events emitted by Node's diagnostic channel TracingChannel https://nodejs.org/api/diagnostics channel.html class-tracingchannel . This is done through @apm-js-collab/tracing-hooks https://github.com/apm-js-collab/tracing-hooks , a helper for orchestrion-js https://github.com/apm-js-collab/orchestrion-js which injects the relevant tracing channels into the instrumented package. We then define a Subscriber that listens to these channels for specific events asyncEnd , asyncStart , start , end , and/or error and record what we need from the event data and context which is… 证据：`lib/subscribers/README.md`
- **Running Benchmarks**（documentation）：The easiest way to run all benchmarks is by using the npm script: 证据：`test/benchmark/Readme.md`
- **Readme**（documentation）：The protobuf schemas in this directory are copied from https://github.com/open-telemetry/opentelemetry-proto/tree/68f9c63/opentelemetry/proto 证据：`test/integration/otel/schemas/Readme.md`
- **Cross Agent Tests**（documentation）：None of these tests should contain customer data such as SQL strings. Please be careful when adding new tests from real world failures. 证据：`test/lib/cross_agent_tests/README.md`
- **CAT Map test details**（documentation）：The CAT map test cases in cat map.json are meant to be used to verify the attributes that agents collect and attach to analytics transaction events for the CAT map project. 证据：`test/lib/cross_agent_tests/cat/README.md`
- **Datastore instance tests**（documentation）：The datastore instance tests provide attributes similar to what an agent could expect to find regarding a database configuration and specifies the expected datastore instance metric https://source.datanerd.us/agents/agent-specs/blob/master/Datastore-Metrics-PORTED.md datastore-metric-namespace that should be generated. The table below lists types attributes and whether they will always be included or optionally included in each test case. 证据：`test/lib/cross_agent_tests/datastores/README.md`
- **Trace Context test details**（documentation）：The Trace Context test cases in trace context.json are meant to be used to verify the creation and forwarding of W3C Trace Context headers within the agent and the attributes and metrics that get created during that process. 证据：`test/lib/cross_agent_tests/distributed_tracing/README.md`
- **Readme**（documentation）：These tests cover parsing of Docker container IDs on Linux hosts out of /proc/self/cgroup or /proc/ /cgroup more generally . 证据：`test/lib/cross_agent_tests/docker_container_id/README.md`
- **Readme**（documentation）：These tests cover parsing of Docker container IDs on Linux hosts out of /proc/self/mountinfo or /proc/ /mountinfo more generally . 证据：`test/lib/cross_agent_tests/docker_container_id_v2/README.md`
- **event source info**（documentation）：This test fixture is intended to verify that the the agent correctly detects event types in languages where type is dynamic and correctly harvests ARN values. The fixtures are structured as follows: 证据：`test/lib/cross_agent_tests/lambda/README.md`
- **PostgreSQL explain plan obfuscation tests**（documentation）：PostgreSQL explain plan obfuscation tests 证据：`test/lib/cross_agent_tests/postgres_explain_obfuscation/README.md`
- **Readme**（documentation）：These tests are for determining the numbers of physical packages, physical cores, and logical processors from the data returned by /proc/cpuinfo on Linux hosts. Each text file in this directory is the output of /proc/cpuinfo on various machines. 证据：`test/lib/cross_agent_tests/proc_cpuinfo/README.md`
- **Readme**（documentation）：These tests are for determining the physical memory from the data returned by /proc/meminfo on Linux hosts. The total physical memory of the linux system is reported as part of the environment values. The key used by the Python agent is 'Total Physical Memory MB '. 证据：`test/lib/cross_agent_tests/proc_meminfo/README.md`
- **Samplers**（documentation）：With the introduction of Otel-style sampling algorithms and core tracing, we now have many configurable samplers that may be working simultaneously within one application. These tests describe how the samplers should be set up based on local config, and how they are expected to behave under traffic. 证据：`test/lib/cross_agent_tests/samplers/README.md`
- **Readme**（documentation）：These test cases cover obfuscation more properly, masking of literal values from SQL statements captured by agents. SQL statements may be captured and attached to transaction trace nodes, or to slow SQL traces. 证据：`test/lib/cross_agent_tests/sql_obfuscation/README.md`
- **Synthetics Tests**（documentation）：The Synthetics tests are designed to verify that the agent handles valid and invalid Synthetics requests. 证据：`test/lib/cross_agent_tests/synthetics/README.md`
- **The Utilization Tests**（documentation）：The Utilization tests ensure that the appropriate information is being gathered for pricing. It is centered around ensuring that the JSON generated by all agents is correct. Each JSON block is a test case, with potentially the following fields: - testname : The name of the test. - input total ram mib : The total ram number calculated by the agent. - input logical processors : The number of logical processors calculated by the agent. - input hostname : The hostname calculated by the agent. - input full hostname : The full hostname calculated by the agent. - input ip address : A string array containing all the values in ip address calculated by the agent. - input aws id : The aws id determine… 证据：`test/lib/cross_agent_tests/utilization/README.md`
- **Vendor Specific Utilization Tests**（documentation）：The Utilization tests ensure that the appropriate information is being gathered for pricing for a particular cloud vendor. It is centered around ensuring that the JSON generated by all agents is correct. Each JSON block is a test case, with potentially the following fields: 证据：`test/lib/cross_agent_tests/utilization_vendor_specific/README.md`
- **Readme**（documentation）：This module is a collection of helper functions to assert expectations in tests. Each file represents a single assertion function, and must export a named function. 证据：`test/lib/custom-assertions/Readme.md`
- **Readme**（documentation）：This folder contains a very basic Express server that is instrumented with the Node.js agent. It will run a test suite that just makes a http requests and verifies that it does not crash the server and properly responds. 证据：`test/smoke/e2e/README.md`
- **Versioned Tests**（documentation）：The subdirectories within this directory represent test suites that verify our agent compatibility across multiple versions of our supported modules. When these tests are run, through our versioned test runner tool, each suite is run multiple times utilizing different versions of the module that satisfy configured version constraints. 证据：`test/versioned/Readme.md`
- **Readme**（documentation）：As of 2024-10-30, the consensus on the team is that this test exists to ensure we are not breaking applications that utilize the continuation-local-storage module. We do not instrument that module, but we shouldn't be breaking it, either. 证据：`test/versioned/cls/Readme.md`
- **Package**（package_manifest）：{ "name": "newrelic", "version": "14.1.0", "author": "New Relic Node.js agent team ", "license": "Apache-2.0", "contributors": { "name": "Saxon D'Aubin", "email": "saxon@newrelic.com", "web": "http://newrelic.com" }, { "name": "Forrest L Norvell", "email": "forrest@newrelic.com", "web": "http://newrelic.com/" }, { "name": "Jacob Groundwater", "email": "jacob@newrelic.com", "web": "https://newrelic.com" }, { "name": "Wraithan Chris McDonald ", "email": "wmcdonald@newrelic.com", "web": "https://newrelic.com" }, { "name": "Michael Hayes", "email": "mhayes@newrelic.com", "web": "https://newrelic.com" }, { "name": "Bryan Clement", "email": "bclement@newrelic.com", "web": "https://newrelic.com" }… 证据：`package.json`
- **Package**（package_manifest）：{ "type": "module", "files": "../../../../index.js" } 证据：`test/integration/config/esm-cjs/package.json`
- **Package**（package_manifest）：{ "type": "module", "files": "../../../../index.js" } 证据：`test/integration/config/esm-js/package.json`
- **Package**（package_manifest）：{ "type": "module", "files": "../../../../index.js" } 证据：`test/integration/config/esm-mjs/package.json`
- **Package**（package_manifest）：{ "name": "uninstrumented-tests", "version": "0.0.0", "private": true, "dependencies": { "amqplib": " ", "cassandra-driver": " ", "connect": " ", "bluebird": " ", "director": " ", "express": " ", "generic-pool": " ", "hapi": " ", "ioredis": " ", "koa": " ", "memcached": " ", "mongodb": " ", "mysql": " ", "mysql2": " ", "pg": " ", "pg.js": " ", "q": " ", "redis": " ", "restify": " ", "superagent": " ", "when": " " }, "engines": { "node": " =22" } } 证据：`test/integration/uninstrumented/package.json`
- **Package**（package_manifest）：{ "name": "@azure/functions-core", "version": "1.0.0", "hostVersion": "1.0.0" } 证据：`test/lib/azure-functions-core/package.json`
- **Package**（package_manifest）：name: "not-json" 证据：`test/lib/example-packages/invalid-json/package.json`
- **Package**（package_manifest）：{ "version": "1.2.3" } 证据：`test/lib/example-packages/valid-json/package.json`
- **Package**（package_manifest）：{ "version": "1.0.0" } 证据：`test/unit/lib/subscribers/fixtures/foo/package.json`
- **Package**（package_manifest）：{ "name": "amqplib-esm-tests", "version": "0.0.0", "type": "module", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "amqplib": { "versions": " =0.5.0", "samples": 1 } }, "files": "issue-2663.test.mjs" } , "dependencies": {} } 证据：`test/versioned/amqplib-esm/package.json`
- **Package**（package_manifest）：{ "name": "ampqlib-tests", "targets": { "name": "amqplib", "minAgentVersion": "2.0.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "amqplib": " =0.5.0" }, "files": "callback.test.js", "promises.test.js" } } 证据：`test/versioned/amqplib/package.json`
- **Package**（package_manifest）：{ "name": "@anthropic-ai/sdk-tests", "targets": { "name": "@anthropic-ai/sdk", "minAgentVersion": "13.19.0" } , "version": "0.0.0", "private": true, "engines": { "node": " =22" }, "tests": { "engines": { "node": " =22" }, "dependencies": { "@anthropic-ai/sdk": " =0.33.0" }, "files": "chat-completions.test.js", "feedback-messages.test.js" } } 证据：`test/versioned/anthropic-sdk/package.json`
- **Package**（package_manifest）：{ "name": "apollo-server-tests", "targets": {"name":"@apollo/gateway","minAgentVersion":"14.0.0"} , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "@apollo/subgraph": "latest", "@apollo/gateway": " =2.3.0", "@apollo/server": "latest", "graphql-tag": "latest" }, "files": "segments.test.js", "service-definition-and-healthcheck-filtering.test.js", "transaction-naming.test.js", "sub-graph-transactions.test.js", "query-obfuscation.test.js" } , "dependencies": {} } 证据：`test/versioned/apollo-federation/package.json`
- **Package**（package_manifest）：{ "name": "apollo-server-tests", "targets": {"name":"@apollo/server","minAgentVersion":"14.0.0"} , "version": "0.0.0", "private": true, "engines": { "node": " =22" }, "tests": { "engines": { "node": " =22" }, "dependencies": { "@apollo/server": " =4.0.0", "graphql-tag": "latest" }, "files": "transaction-naming.test.js", "segments.test.js", "agent-disabled.test.js", "errors.test.js", "attributes.test.js", "query-obfuscation.test.js", "metrics.test.js" } , "dependencies": {} } 证据：`test/versioned/apollo-server/package.json`
- **Package**（package_manifest）：{ "name": "aws-sdk-v2-tests", "version": "0.0.0", "targets": {"name":"aws-sdk","minAgentVersion": "6.2.0"} , "private": true, "tests": { "engines": { "node": " =22.0" }, "dependencies": { "aws-sdk": { "versions": "2.2.48" } }, "files": "instrumentation-unsupported.test.js" }, { "engines": { "node": " =22.0" }, "dependencies": { "aws-sdk": { "versions": " =2.463.0", "samples": 1 } }, "files": "aws-sdk.test.js", "dynamodb.test.js", "http-services.test.js", "instrumentation-supported.test.js", "s3.test.js", "sns.test.js", "sqs.test.js" }, { "engines": { "node": " =22.0" }, "dependencies": { "amazon-dax-client": { "versions": " =1.2.5", "samples": 1 } }, "files": "amazon-dax-client.test.js" } ,… 证据：`test/versioned/aws-sdk-v2/package.json`
- **Package**（package_manifest）：{ "name": "aws-sdk-v3-tests", "engines": { "node": " =22" }, "targets": { "name": "@aws-sdk/client-sqs", "minAgentVersion": "8.7.1" }, { "name": "@aws-sdk/client-sns", "minAgentVersion": "8.7.1" }, { "name": "@aws-sdk/client-dynamodb", "minAgentVersion": "8.7.1" }, { "name": "@aws-sdk/lib-dynamodb", "minAgentVersion": "8.7.1" }, { "name": "@aws-sdk/smithy-client", "minSupported": "3.47.0", "minAgentVersion": "8.7.1" }, { "name": "@smithy/smithy-client", "minSupported": "2.0.0", "minAgentVersion": "11.0.0" }, { "name": "@aws-sdk/client-bedrock-runtime", "minAgentVersion": "11.13.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "comment": "These are sample… 证据：`test/versioned/aws-sdk-v3/package.json`
- **Package**（package_manifest）：{ "name": "azure-functions-tests", "targets": { "name": "@azure/functions", "minAgentVersion": "12.18.0", "minSupported": "4.7.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "@azure/functions": " =4.7.0" }, "files": "background.test.js", "http.test.js", "logs.test.js" } } 证据：`test/versioned/azure-functions/package.json`
- **Package**（package_manifest）：{ "name": "bluebird-tests", "targets": {"name": "bluebird", "minAgentVersion": "1.27.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "bluebird": " =3.0.0" }, "files": "baseline.test.js", "promise-all.test.js", "promise-all-settled.test.js", "promise-any.test.js", "promise-bind.test.js", "promise-call.test.js", "promise-catch-return.test.js", "promise-catch-throw.test.js", "promise-coroutine.test.js", "promise-constructor.test.js", "promise-delay.test.js", "promise-each.test.js", "promise-error.test.js", "promise-filter.test.js", "promise-get.test.js", "promise-get-new-library-copy.test.js", "promise-is.test.js", "promise-join.test.js",… 证据：`test/versioned/bluebird/package.json`
- **Package**（package_manifest）：{ "name": "bunyan-tests", "targets": {"name":"bunyan", "minAgentVersion": "9.3.0"} , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "bunyan": " =1.8.12" }, "files": "bunyan.test.js" } } 证据：`test/versioned/bunyan/package.json`
- **Package**（package_manifest）：{ "name": "cassandra-driver-tests", "targets": {"name":"cassandra-driver","minAgentVersion": "1.7.1"} , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22 =4.0.0" }, "files": "query.test.js" }, { "engines": { "node": " =24" }, "dependencies": { "cassandra-driver": " =4.8.0" }, "files": "query.test.js" } } 证据：`test/versioned/cassandra-driver/package.json`
- **Package**（package_manifest）：{ "name": "cls-tests", "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "continuation-local-storage": " =3.1.2" }, "files": "cls.test.js" } } 证据：`test/versioned/cls/package.json`
- **Package**（package_manifest）：{ "name": "connect-tests", "targets": {"name":"connect","minAgentVersion":"2.6.0"} , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "connect": " =3.0.0" }, "files": "error-intercept.test.js", "route.test.js" } } 证据：`test/versioned/connect/package.json`
- **Package**（package_manifest）：{ "name": "disabled-instrumentation-tests", "targets": , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "express": "latest", "ioredis": "latest", "mongodb": "latest" }, "files": "disabled-express.test.js", "disabled-ioredis.test.js" } } 证据：`test/versioned/disabled-instrumentation/package.json`
- **Package**（package_manifest）：{ "name": "elasticsearch-tests", "targets": {"name":"@elastic/elasticsearch","minAgentVersion":"11.9.0"} , "version": "0.0.0", "private": true, "engines": { "node": " =22" }, "tests": { "supported": false, "comment": "Used to assert our instrumentation does not get loaded on old versions.", "engines": { "node": " =22" }, "dependencies": { "@elastic/elasticsearch": "7.13.0" }, "files": "elasticsearchNoop.test.js" }, { "engines": { "node": " =22" }, "dependencies": { "@elastic/elasticsearch": " =7.16.0" }, "files": "elasticsearch.test.js" } } 证据：`test/versioned/elastic/package.json`
- **Package**（package_manifest）：{ "name": "esm-package-tests", "description": "Integration test our ESM Loader hooks with an ESM package from the ecosystem", "version": "0.0.0", "type": "module", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "parse-json": { "versions": "6.0.2", "samples": 1 } }, "files": "parse-json.test.mjs" } , "dependencies": {} } 证据：`test/versioned/esm-package/package.json`
- **Package**（package_manifest）：{ "name": "express-esm-tests", "targets": {"name":"express","minAgentVersion":"2.6.0"} , "version": "0.0.0", "type": "module", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "express": { "versions": " =4.15.0", "samples": 1 } }, "files": "segments.test.mjs", "transaction-naming.test.mjs" } } 证据：`test/versioned/express-esm/package.json`
- **Package**（package_manifest）：{ "name": "express-tests", "targets": { "name": "express", "minAgentVersion": "2.6.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "express": { "versions": " =4.15.0", "samples": 5 }, "express-enrouten": "1.1", "ejs": "2.5.9" }, "files": "app-use.test.js", "async-error.test.js", "async-handlers.test.js", "bare-router.test.js", "captures-params.test.js", "client-disconnect.test.js", "errors.test.js", "express-enrouten.test.js", "ignoring.test.js", "issue171.test.js", "middleware-name.test.js", "render.test.js", "require.test.js", "route-iteration.test.js", "route-param.test.js", "router-params.test.js", "segments.test.js", "transaction-… 证据：`test/versioned/express/package.json`
- **Package**（package_manifest）：{ "name": "fastify-tests", "targets": { "name": "fastify", "minAgentVersion": "8.5.0" } , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "supported": false, "comment": "Used to assert our instrumentation does not get loaded on old versions.", "dependencies": { "fastify": "3.29.0", "middie": "3.2.1" }, "files": "fastify-no-op.test.js" }, 证据：`test/versioned/fastify/package.json`
- **Package**（package_manifest）：{ "name": "generic-pool-tests", "targets": {"name":"generic-pool","minAgentVersion":"0.9.0"} , "version": "0.0.0", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "generic-pool": " =3.0.0" }, "files": "basic.test.js" } } 证据：`test/versioned/generic-pool/package.json`
- **Package**（package_manifest）：{ "name": "google-adk-tests", "targets": { "name": "@google/adk", "minSupported": "1.1.0", "minAgentVersion": "13.20.0" } , "version": "0.0.0", "private": true, "engines": { "node": " =22" }, "tests": { "engines": { "node": " =22" }, "dependencies": { "@google/adk": " =1.1.0" }, "files": "agent-run-async.test.js", "tool-run-async.test.js" } } 证据：`test/versioned/google-adk/package.json`
- **Package**（package_manifest）：{ "name": "@google/genai-tests", "targets": { "name": "@google/genai", "minAgentVersion": "12.21.0" } , "version": "0.0.0", "private": true, "engines": { "node": " =22" }, "tests": { "engines": { "node": " =22" }, "dependencies": { "@google/genai": " =1.1.0 =1.5.1" }, "files": "chat-completions.test.js", "embeddings.test.js", "feedback-messages.test.js" } } 证据：`test/versioned/google-genai/package.json`
- **Package**（package_manifest）：{ "name": "grpc-esm-tests", "targets": {"name":"@grpc/grpc-js","minAgentVersion":"8.17.0"} , "version": "0.0.0", "type": "module", "private": true, "tests": { "engines": { "node": " =22" }, "dependencies": { "@grpc/grpc-js": { "versions": " =1.4.0", "samples": 1 } }, "files": "client-unary.test.mjs", "server-unary.test.mjs" } } 证据：`test/versioned/grpc-esm/package.json`
- 其余 20 条证据见 `AI_CONTEXT_PACK.json` 或 `EVIDENCE_INDEX.json`。

## 宿主 AI 必须遵守的规则

- **把本资产当作开工前上下文，而不是运行环境。**：AI Context Pack 只包含证据化项目理解，不包含目标项目的可执行状态。 证据：`README.md`, `cloud-tooling/README.md`, `docker/Readme.md`
- **回答用户时区分可预览内容与必须安装后才能验证的内容。**：安装前体验的消费者价值来自降低误装和误判，而不是伪装成真实运行。 证据：`README.md`, `cloud-tooling/README.md`, `docker/Readme.md`

## 用户开工前应该回答的问题

- 你准备在哪个宿主 AI 或本地环境中使用它？
- 你只是想先体验工作流，还是准备真实安装？
- 你最在意的是安装成本、输出质量、还是和现有规则的冲突？

## 验收标准

- 所有能力声明都能回指到 evidence_refs 中的文件路径。
- AI_CONTEXT_PACK.md 没有把预览包装成真实运行。
- 用户能在 3 分钟内看懂适合谁、能做什么、如何开始和风险边界。

---

## Doramagic Context Augmentation

下面内容用于强化 Repomix/AI Context Pack 主体。Human Manual 只提供阅读骨架；踩坑日志会被转成宿主 AI 必须遵守的工作约束。

## Human Manual 骨架

使用规则：这里只是项目阅读路线和显著性信号，不是事实权威。具体事实仍必须回到 repo evidence / Claim Graph。

宿主 AI 硬性规则：
- 不得把页标题、章节顺序、摘要或 importance 当作项目事实证据。
- 解释 Human Manual 骨架时，必须明确说它只是阅读路线/显著性信号。
- 能力、安装、兼容性、运行状态和风险判断必须引用 repo evidence、source path 或 Claim Graph。

- **Agent 核心架构与数据流**：importance `high`
  - source_paths: index.js, newrelic.js, lib/agent.js, lib/harvester.js, lib/collector/api.js
- **插桩(Instrumentation)与订阅者(Subscribers)机制**：importance `high`
  - source_paths: lib/shimmer.js, lib/shim/index.js, lib/shim/datastore-shim.js, lib/shim/webframework-shim/index.js, lib/shim/message-shim/index.js
- **配置系统、ESM 支持与自定义扩展**：importance `high`
  - source_paths: newrelic.js, api.js, stub_api.js, esm-loader.mjs, lib/config/default.js
- **AI/LLM 监控、Next.js 混合 Agent 与 OpenTelemetry 集成**：importance `high`
  - source_paths: lib/instrumentation/nextjs/next-server.js, lib/instrumentation/nextjs/nr-hooks.js, lib/instrumentation/nextjs/utils.js, lib/otel/setup.js, lib/otel/traces/index.js

## Repo Inspection Evidence / 源码检查证据

- repo_clone_verified: true
- repo_inspection_verified: true
- repo_commit: `185bb66c6b8c8ddc26c934adccc28bf47e9ddfd9`
- inspected_files: `README.md`, `docker-compose.yml`, `package.json`, `examples/instrumentation.md`, `examples/jsdoc.json`

宿主 AI 硬性规则：
- 没有 repo_clone_verified=true 时，不得声称已经读过源码。
- 没有 repo_inspection_verified=true 时，不得把 README/docs/package 文件判断写成事实。
- 没有 quick_start_verified=true 时，不得声称 Quick Start 已跑通。

## Doramagic Pitfall Constraints / 踩坑约束

这些规则来自 Doramagic 发现、验证或编译过程中的项目专属坑点。宿主 AI 必须把它们当作工作约束，而不是普通说明文字。

### Constraint 1: 来源证据：Drop support for next < 16

- Trigger: GitHub 社区证据显示该项目存在一个维护/版本相关的待验证问题：Drop support for next < 16
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4054 | 来源讨论提到 node 相关条件，需在安装/试用前复核。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 2: 来源证据：Drop support for prisma wasm client

- Trigger: GitHub 社区证据显示该项目存在一个维护/版本相关的待验证问题：Drop support for prisma wasm client
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4053 | 来源类型 github_issue 暴露的待验证使用条件。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 3: 来源证据：Drop support for restify

- Trigger: GitHub 社区证据显示该项目存在一个维护/版本相关的待验证问题：Drop support for restify
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4055 | 来源讨论提到 node 相关条件，需在安装/试用前复核。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 4: 来源证据：[spike] Research migrating security agent to use tracing channel instrumentation

- Trigger: GitHub 社区证据显示该项目存在一个安全/权限相关的待验证问题：[spike] Research migrating security agent to use tracing channel instrumentation
- Why it matters: 可能影响升级、迁移或版本选择。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4056 | 来源讨论提到 node 相关条件，需在安装/试用前复核。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 5: 仓库名和安装名不一致

- Trigger: 仓库名 `node-newrelic` 与安装入口 `newrelic` 不完全一致。
- Host AI rule: 在 npm/PyPI/GitHub 上确认包名映射和官方 README 说明。
- Why it matters: 用户照着仓库名搜索包或照着包名找仓库时容易走错入口。
- Evidence: identity.distribution | https://www.npmjs.com/package/newrelic | repo=node-newrelic; install=newrelic
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 6: 来源证据：Fix Next.js tests

- Trigger: GitHub 社区证据显示该项目存在一个配置相关的待验证问题：Fix Next.js tests
- Host AI rule: 来源显示可能已有修复、规避或版本变化，说明书中必须标注适用版本。
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/3458 | 来源类型 github_issue 暴露的待验证使用条件。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 7: 能力判断依赖假设

- Trigger: README/documentation is current enough for a first validation pass.
- Host AI rule: 将假设转成下游验证清单。
- Why it matters: 假设不成立时，用户拿不到承诺的能力。
- Evidence: capability.assumptions | https://www.npmjs.com/package/newrelic | README/documentation is current enough for a first validation pass.
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 8: 来源证据：Add supportability metrics to OTEL metrics exporter

- Trigger: GitHub 社区证据显示该项目存在一个运行相关的待验证问题：Add supportability metrics to OTEL metrics exporter
- Host AI rule: 来源显示可能已有修复、规避或版本变化，说明书中必须标注适用版本。
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4050 | 来源讨论提到 node 相关条件，需在安装/试用前复核。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 9: 来源证据：[spike] Research replacing instrument* api methods

- Trigger: GitHub 社区证据显示该项目存在一个维护/版本相关的待验证问题：[spike] Research replacing instrument* api methods
- Why it matters: 可能影响升级、迁移或版本选择。
- Evidence: community_evidence:github | https://github.com/newrelic/node-newrelic/issues/4057 | 来源讨论提到 node 相关条件，需在安装/试用前复核。
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 10: 维护活跃度未知

- Trigger: 未记录 last_activity_observed。
- Host AI rule: 补 GitHub 最近 commit、release、issue/PR 响应信号。
- Why it matters: 新项目、停更项目和活跃项目会被混在一起，推荐信任度下降。
- Evidence: evidence.maintainer_signals | https://www.npmjs.com/package/newrelic | last_activity_observed missing
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。