Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add flanliulf/SpecLite --skill speclite-brownfield-context-buildergit clone --depth 1 https://github.com/flanliulf/SpecLiteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/flanliulf/speclite/speclite-brownfield-context-builder)<a href="https://agentmods.dev/skills/flanliulf/speclite/speclite-brownfield-context-builder"><img src="https://agentmods.dev/badge/skills/flanliulf/speclite/speclite-brownfield-context-builder/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/flanliulf/speclite/speclite-brownfield-context-builder"><img src="https://agentmods.dev/badge/skills/flanliulf/speclite/speclite-brownfield-context-builder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00077 | $0.01837 |
| Opus 5 | $0.00039 | $0.00919 |
| Sonnet 5 | $0.00015 | $0.00367 |
| Haiku 4.5 | $0.00008 | $0.00184 |
Grade A, and why
speclite-brownfield-context-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
[技能说明] Speclite Brownfield Context Builder 是 1-analysis 阶段的既有系统分析工作流。它把现有代码仓库恢复为可追溯、可复查、可供后续 PRD / Architecture / Epics / Stories 消费的 as-is 基线和规划前置上下文。
[核心能力]
- 仓库分类与边界识别:识别 monolith / monorepo / multi-part 结构,划分子系统边界与技术栈。
- 确定性证据提取:通过 {skill-root}/scripts/ 下的脚本提取 API、数据模型、依赖图、配置面、测试面和历史文档候选事实。
- Evidence-First 基线合成:先生成 evidence 与 baseline skeleton,再由 LLM 仅补受限叙述,避免无来源事实写入。
- 历史文档交叉验证:将历史 PRD / 技术方案作为候选事实,经过代码或当前文档验证后再进入基线。
- 局部深潜分析:对高风险或用户指定区域做 targeted deep dive,补充可复用能力、约束和变更风险。
- 规划前置交接:默认生成 brownfield-planning-brief、candidate-change-slices、feature-entry-points 等后续 Speclite 工作流输入,不替代 create-prd / create-architecture / create-epics-and-stories。
- 状态持久化与恢复:每阶段写盘并更新 project-scan-report.json,支持 resume 和 full rescan 归档。
- 质量护栏:执行 coverage contract、strict anchors、adversarial grounding check 和 golden regression。
[约定]
裸路径相对于 {skill-root} 解析;{project-root} 是目标项目工作目录;{speclite-runtime-root} 是 {project-root}/_speclite;{skill-name} 是目录 basename。
运行配置必须运行 `speclite resolve config --project-root {project-root}`。推荐解析字段:
- `{project_knowledge}` = `[modules.sdlc].project_knowledge`
- `{planning_artifacts}` = `[modules.sdlc].planning_artifacts`
- `{brownfield_output}` = `{project_knowledge}/brownfield`
- `{history_sources}` 默认 `{project-root}/docs/history`,可由用户指定
[执行流程]
本 Skill 采用 4 种运行模式和 6 个阶段,路由规则以 references/workflow-router.md 为准。
1. 激活后解析三层 customize:`{skill-root}/customize.toml`、`{speclite-runtime-root}/custom/{skill-name}.toml`、`{speclite-runtime-root}/custom/{skill-name}.user.toml`。缺失覆盖文件时跳过;标量覆盖,表深度合并。
2. 运行 `speclite resolve config --project-root {project-root}`。配置缺失或 `project_knowledge`、`planning_artifacts` 为空时 HALT。
3. Phase 0:检查 `{brownfield_output}/project-scan-report.json`,按 `initial_scan` / `full_rescan` / `targeted_deep_dive` / `planning_generation` / resume 判定执行路径。初始化状态时运行:`python {skill-root}/scripts/update_state.py init --project-root {project-root} --output-dir {brownfield_output}`。
4. Phase 1:运行 `python {skill-root}/scripts/scan_repo.py {project-root} --output-dir {brownfield_output}`,并按 `references/repository-classifier.md` 划分仓库结构。
5. Phase 2:按 `references/evidence-extractor.md` 执行证据提取脚本,所有脚本必须显式传入 `--output-dir {brownfield_output}`;历史文档摄取额外传入 `--history-dir {history_sources}`。完成后运行 `validate_outputs.py --phase evidence`。
6. Phase 3:读取 `references/known-failure-patterns.md`,运行 `render_baseline_skeleton.py` 生成 skeleton,再按 `references/baseline-synthesizer.md` 补充受限叙述。完成后必须运行 `adversarial_grounding_check.py` 和 `validate_outputs.py --phase baseline --strict-anchors`。
7. Phase 4:如用户指定目标区域,按 `references/deep-dive-analyzer.md` 生成 `deep-dives/deep-dive-{area}.md`,并将发现反馈到 baseline 或 planning brief。
8. Phase 5:按 `references/planning-synthesizer.md` 生成规划前置产物。默认输出 `brownfield-planning-brief.md`、`candidate-change-slices.md`、`feature-entry-points.md` 到 `{planning_artifacts}` 或 `{brownfield_output}/planning/`,并明确后续应交给 Speclite `speclite-create-prd`、`speclite-create-architecture`、`speclite-create-epics-and-stories` 继续处理。
[输出目录结构]
默认产物写入目标项目的 {project_knowledge}/brownfield/,规划前置交接文档可同步到 {planning_artifacts}:
```text
{project_knowledge}/brownfield/
├── project-scan-report.json
├── evidence/
├── baseline/
├── deep-dives/
├── planning/
└── validation/
```
[核心边界原则] 1. 不伪造历史:不反推研发历史,不虚构优先级、排期或验收过程。 2. 历史文档是证据源,不是真相源:历史资料必须经过候选事实、交叉验证、基线事实路径。 3. as-is 与 to-be 分离:baseline 只写当前系统事实,planning 只写后续规划输入。 4. 不吞并后续阶段:本 Skill 可生成 planning brief 和候选切片,但默认不替代 Speclite PRD、Architecture、Epics/Stories 工作流。 5. 强落盘、弱上下文:阶段完成后立即写盘并更新状态,只在上下文中保留摘要。 6. 证据优先于叙述:每个关键结论必须有 evidence anchor;证据不足时标注不确定性,不补全事实。
[注意事项]
- 所有脚本调用必须使用 {skill-root}/scripts/...,不得依赖源码仓库路径。
- 所有运行输出必须使用 {brownfield_output} 或 {planning_artifacts},不得把 {project_knowledge}/brownfield 写成当前运行依赖。
- config.toml.example 仅作字段结构参考,不作为 runtime fallback。
- 修改 scripts/extract_*.py、framework adapters 或 skeleton renderer 后,必须运行 python {skill-root}/scripts/run_golden.py {skill-root}/golden。
- 事实冲突(fact-conflicts.json)必须提示用户确认,不得自动裁决。
- 输出文档末尾必须追加 *本文档由 speclite-brownfield-context-builder Skill 自动生成* 标注。
What ships with it
40 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/architecture-template.md 1.4 KB
- assets/brownfield-planning-brief-template.md 1.5 KB
- assets/epics-template.md 861 B
- assets/prd-template.md 1.1 KB
- assets/story-template.md 1.7 KB
- CHANGELOG.md 5.2 KB
- config.toml.example 485 B
- customize.toml 471 B
- golden/mybatis-plus-basic/dao/entity/UserInfo.java 326 B
- golden/mybatis-plus-basic/expected/data-model-inventory.expected.json 282 B
- golden/README.md 984 B
- golden/spring-mvc-basic/expected/api-inventory.expected.json 399 B
- golden/spring-mvc-basic/src/main/java/com/example/lucky/LuckyController.java 429 B
- references/baseline-synthesizer.md 10 KB
- references/deep-dive-analyzer.md 7.7 KB
- references/document-templates.md 4.2 KB
- references/evidence-extractor.md 6.2 KB
- references/evidence-schema.md 10 KB
- references/examples.md 5.2 KB
- references/framework-adapters/mybatis-plus.md 2.9 KB
- references/framework-adapters/spring-mvc.md 3.5 KB
- references/historical-doc-ingestion.md 5.8 KB
- references/known-failure-patterns.md 7.9 KB
- references/planning-synthesizer.md 9.4 KB
- references/repository-classifier.md 5.1 KB
- references/v2-paradigm-design.md 5.8 KB
- references/validation-checklist.md 3.8 KB
- references/workflow-router.md 6.5 KB
- scripts/adversarial_grounding_check.py 11 KB runs code
- scripts/build_dependency_graph.py 11 KB runs code
- scripts/collect_config_surface.py 13 KB runs code
- scripts/collect_test_surface.py 5.3 KB runs code
- scripts/extract_api_inventory.py 13 KB runs code
- scripts/extract_data_models.py 16 KB runs code
- scripts/merge_historical_facts.py 9.1 KB runs code
- scripts/render_baseline_skeleton.py 8.3 KB runs code
- scripts/run_golden.py 5.1 KB runs code
- scripts/scan_repo.py 16 KB runs code
- scripts/update_state.py 11 KB runs code
- scripts/validate_outputs.py 17 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 83 lines · 77 tokens per session scan A 0893858d39f3
speclite-brownfield-context-builder is a skill published in the GitHub repository flanliulf/SpecLite (4 stars, last pushed 2mo ago), licensed MIT. It adds 77 tokens to every session and 1,837 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
lain
Structural code intelligence for AI coding agents. Use this skill when the user wants to understand how a codebase is organized (modules, call graphs, file dependencies), find where to start reading, trace the impact of a change, find code by meaning, or understand what a symbol does in its full context. Do NOT use…
botpipe-workflow-authoring
Author, review, and improve Botpipe workflows. Use when Codex is asked to create packaged or workspace-local Botpipe workflows, convert codebases into workflows, design provider-heavy producer/verifier steps, write Botpipe prompts/contracts, inspect Botpipe traces, or apply Codex CLI/gpt-5.5 workflow patterns.
intake
The planning front-door. Captures feature ideas and bugs as graded rows in the root INTAKE.md ledger. Use it when the user says "log an idea", "capture a bug", "add to the backlog", "note this down", "track this feature", or invokes /intake. Owns the requirements interview — fleshes out thin ideas, proactively…
merge
The ship gate — the only skill that merges. --staging merges the feature→staging PR on green CI, deploys, verifies, emits a human test script and stamps the staging sign-off on approval. --production ships the double-confirmed release to main and deploys production. Never self-certifies staging; nothing reaches main…
pre-merge
The CI gate. Takes a feature branch from "eng says done" to "PR open against staging with green checks". Runs the project's preflight-resolved pipeline from devkit/policy.json components[]: sync → parallel correctness + security waves → coverage → regression tail → security/migration → PRD-consistency → open PR. Emits…
msg
Root menu for msg skills, plus harness modes. --init is the one-time project bootstrap — use it when the user says "initialise project", "bootstrap repo", "set up the framework", "start a new project", or asks to set up project structure in an empty repo. Other modes: --init-staging (add a staging branch), --update…