Agent skill

OpenCLI Adapter Author

by jackwener in jackwener/OpenCLI

Walks through writing an OpenCLI adapter for a new site or a new command on an existing one, from first recon and field decoding to coding and verification.

Apache-2.0Auto-check passedDevelopment

SKILL.md written in Chinese; this summary is our English description.

Install OpenCLI Adapter Author

skills CLI
$ npx skills add jackwener/OpenCLI --skill opencli-adapter-author -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jackwener/OpenCLI opencli-adapter-author --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jackwener/OpenCLI.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/opencli-adapter-author .claude/skills/opencli-adapter-author && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
opencli-adapter-author
GitHub stars
30k
Used in
1 other repo
Token cost
~3.4k tokens
SKILL.md length
605 words
Files
19 (incl. references)
Skills in repo
5
Repo updated
First seen
Licence
Apache-2.0

At a glance

Walks through writing an OpenCLI adapter for a new site or a new command on an existing one, from first recon and field decoding to coding and verification.

  • Works in 3 steps: 数据在浏览器里看得到吗?(否 → 先解决鉴权) → 数据是 HTTP/JSON/HTML 吗?(否 → 不在 skill 范围) → 需要实时推送吗?(是 → 找同数据 HTTP 接口;没有就放弃)
  • Writing an OpenCLI adapter for a new website
  • SKILL.md covers 前置:看你落在哪, 顶层决策树, Runbook(一步一步勾选) and 降级路径(某步卡住跳到哪), plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

The goal is for a simple site to go from nothing to a passing `opencli browser verify` within about 30 minutes, while complex, private-protocol or write-operation sites put complete evidence and safety ahead of the time limit. The agent uses only existing commands (`opencli browser`, `opencli doctor`, `opencli browser init`, `opencli browser verify`) and debugs browser adapters with tracing on and the tab kept in the foreground. A quick self-test from a coverage matrix asks whether the data is visible in the browser, is HTTP, JSON or HTML, and needs no real-time push.

Before writing code the agent records a strategy note choosing among six approaches: a public API, a cookie-authenticated API, a page-context fetch, request interception, state embedded in the page, or UI selectors. The choice depends on whether the data source has an external contract: official interfaces are the steadiest, visible UI is next, and undocumented internal calls drift most. The skill reuses only data the page legitimately obtains and does not teach breaking signatures or bypassing captchas or access control. References cover recon, API discovery, field decoding, output design, typed errors, fixtures and notes on specific sites. The skill text is mostly in Chinese.

When your agent uses it

  • Writing an OpenCLI adapter for a new website
  • Adding another command to an existing site adapter
  • Choosing between a public API, cookie API and page scraping for a site

Example prompts

  • “Write an OpenCLI adapter for the Xueqiu hot topics page and verify it.”
  • “Add a new search command to the existing bilibili adapter.”
  • “Do recon on this site and tell me which strategy is the most stable before writing code.”

Requirements

  • The `opencli` CLI with a logged-in browser session
  • `opencli doctor` reporting a healthy setup
  • Pre-approved tools (allowed-tools): Bash(opencli:*), Bash(jsluice:*), Read, Edit, Write, Grep

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. 数据在浏览器里看得到吗?(否 → 先解决鉴权)
  2. 数据是 HTTP/JSON/HTML 吗?(否 → 不在 skill 范围)
  3. 需要实时推送吗?(是 → 找同数据 HTTP 接口;没有就放弃)

What it can do on your machine

Read from SKILL.md and the folder at commit 2413694. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Bash(opencli:*)
    • Bash(jsluice:*)
    • Read
    • Edit
    • Write
    • Grep

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

OpenCLI Adapter Author loads about 3.4k tokens when it runs, and up to ~39k if it reads all its reference files. Until then it costs about 95 tokens; SKILL.md has 605 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~95
When it runs · the whole SKILL.md, loaded when a task matches
~3.4k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~39k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jackwener/OpenCLI at commit 2413694, republished under its Apache-2.0 licence (© jackwener). 605 words, ~3,435 tokens.

Download SKILL.mdSave it as .claude/skills/opencli-adapter-author/SKILL.md (or your agent's skills folder). This skill also uses 18 other files; get the full folder from GitHub.
name
opencli-adapter-author
description
Use when writing an OpenCLI adapter for a new site or adding a new command to an existing site. Guides end-to-end from first recon through field decoding, adapter coding, and verify. Replaces opencli-oneshot / opencli-explorer. For ad-hoc browser driving (no adapter), see opencli-browser instead; for a top-level orientation to opencli, see opencli-usage.
allowed-tools
Bash(opencli:*), Bash(jsluice:*), Read, Edit, Write, Grep

opencli-adapter-author

你是要给一个站点写 adapter 的 agent。这份 skill 目标:简单站点争取 30 分钟内从零到通过 opencli browser verify;复杂、私有协议或写操作站点以证据完整和安全为先,不为了时限猜接口。

全程用现有工具:opencli browser * / opencli doctor / opencli browser init / opencli browser verify。没有新命令。

调试浏览器型 adapter 时,优先直接带上 --trace on --keep-tab true --window foreground。--trace on 每轮都落 trace artifact,summary.md 是失败/成功复盘入口;--keep-tab true --window foreground 让 tab lease 保留且浏览器窗口在前台,方便核对最终页面状态。


前置:看你落在哪

先拿 coverage-matrix.md 快速自测。三个问题:

  1. 数据在浏览器里看得到吗?(否 → 先解决鉴权)
  2. 数据是 HTTP/JSON/HTML 吗?(否 → 不在 skill 范围)
  3. 需要实时推送吗?(是 → 找同数据 HTTP 接口;没有就放弃)

三个都 yes 继续。


顶层决策树

先定 strategy,再写 adapter。 每次进入 Step 3/4 后、写代码前,必须产出一段 strategy note。没有这段 note,不要开始写 clis/<site>/<name>.js。

核心判断不是 "API 比 DOM 高级",而是 数据源有没有外部契约。实测维护成本显示:公开/官方接口最稳;UI/DOM 语义通常也有用户可见契约;站内未文档化 XHR/GraphQL/signature endpoint 最容易漂。不要为了 "API-first" 把稳定的 UI/DOM 实现盲目迁到无契约内部接口。

md
Strategy: PUBLIC_API | COOKIE_API | PAGE_FETCH | INTERCEPT | DOM_STATE | UI_SELECTOR
Contract: stable | visible-ui | internal-unstable
Evidence:
- observed request/state: <endpoint / state global / UI-only signal>
- auth source: <none / browser cookie / csrf from meta / localStorage / page runtime>
- replay result: <status + content-type + non-empty sample shape>

If Strategy is PAGE_FETCH or INTERCEPT:
- why PUBLIC_API / COOKIE_API are unavailable:
- why UI_SELECTOR / DOM_STATE are not safer:
- why the maintenance cost is acceptable:

Strategy classes:

Strategy契约级别用在什么时候证据要求
PUBLIC_APIstable不需要登录,Node-side fetch 直接拿到目标数据200 + JSON/HTML 含目标数据,不是埋点/广告
COOKIE_APIstableNode-side fetch + page.getCookies() / header helper 能拿数据cookie/CSRF 来源清楚,replay 非空
UI_SELECTORvisible-uipublish/upload/click/表单,或页面语义比内部接口更稳selector 有语义锚点;错误路径是 typed error
DOM_STATEvisible-ui数据在 hydration state / bootstrap JSON / SSR HTML 里state key / script JSON / HTML 结构明确
PAGE_FETCHinternal-unstable只能在页面上下文 fetch 才能复用 same-origin/session/runtimeopencli browser eval fetch(...) 非空;必须解释为什么避不开内部接口
INTERCEPTinternal-unstable请求签名复杂,但页面自己能自然发出请求触发 UI 后能截到目标 response;必须解释为什么 UI/DOM 不够

选择规则:优先 PUBLIC_API / COOKIE_API。如果 UI/DOM 语义稳定,不要强行升级到 PAGE_FETCH / INTERCEPT。只有公开/官方接口不可用、UI/DOM 无法表达目标数据或操作时,才承担无契约内部接口的维护成本。

实测:PAGE_FETCH / INTERCEPT 的 fix 频率约为 PUBLIC_API 的 7-8 倍,UI_SELECTOR 跟 COOKIE_API 同档。详细 ladder 推导、api_candidates 证据怎么填、booking #1680 等反例见 references/strategy-selection.md。

边界:只复用页面自己已经合法获得的数据/能力。不教破解签名、不绕验证码/风控/访问控制;遇到不可复用签名(如必须由页面 runtime 生成且不能安全抽象)就降级到 UI_SELECTOR / DOM_STATE / INTERCEPT。

START
  │
  ▼
┌──────────────────────────┐
│ opencli doctor 通?      │── no ──→ 修桥接(doctor 输出里的提示)
└──────────────────────────┘
  │ yes
  ▼
┌────────────────────────────────────────────────────┐
│ 读站点记忆:                                        │
│   1. ~/.opencli/sites/<site>/endpoints.json         │
│   2. ~/.opencli/sites/<site>/notes.md               │
│   3. references/site-memory/<site>.md               │
└────────────────────────────────────────────────────┘
  │ 命中 endpoint + 字段 → 直接跳到【endpoint 验证】(不跳写 adapter!memory 可能过期)
  │ 没命中 → 继续
  ▼
┌──────────────────────────┐
│ 站点侦察(site-recon)    │  → Pattern A/B/C/D/E
└──────────────────────────┘
  │
  ▼
┌──────────────────────────┐
│ API 发现(api-discovery)│  §1 network → §2 state → §3 bundle → §4 token → §5 intercept
└──────────────────────────┘
  │ 拿到候选 endpoint
  ▼
┌────────────────────────────────────────────┐
│ 需要 Deep Recon?                          │  无文档私有 API / DOM 丢数据 / 写操作 / 证据冲突
│ → references/deep-recon.md                 │  intent matrix → 因果 diff → 候选账本 → contract gate
└────────────────────────────────────────────┘
  │ 候选通过合同证明;不通过则记录拒绝与 lift condition
  ▼
┌────────────────────────────────────────────┐
│ 验证候选合同(memory 命中也要跑)           │── 401/403 ──→ 回到 §4 排 token
│ safe replay;不可 replay 的 read 用自然截获 │── 空/HTML ──→ 回到 site-recon 换 Pattern
│ 数据非空、identity 对、分页/错误语义完整     │── 站点换版 ──→ 标记旧 endpoint,回 api-discovery
└────────────────────────────────────────────┘
  │ OK
  ▼
┌───────────────────────────────────────┐
│ 字段解码(memory 里的 field-map 也要抽查)│  自解释 → 直接 / 已知代号 → field-conventions / 未知 → decode-playbook
│ 比一条已知字段和网页肉眼值,确认没错位     │
└───────────────────────────────────────┘
  │
  ▼
┌──────────────────────────┐
│ 设计 columns (output)    │  对照 output-design.md 的命名 / 类型 / 顺序
└──────────────────────────┘
  │
  ▼
┌──────────────────────────┐
│ opencli browser init      │  生成 ~/.opencli/clis/<site>/<name>.js 骨架
│ 复制最像的邻居 adapter    │
│ 改 name / URL / 映射三处  │
└──────────────────────────┘
  │
  ▼
┌──────────────────────────┐
│ opencli browser verify    │── 失败 ──→ autofix skill,用 --trace retain-on-failure 回对应步骤
└──────────────────────────┘
  │ 成功
  ▼
┌──────────────────────────┐
│ 字段 vs 网页肉眼对一遍   │── 数值不对 ──→ 回字段解码
└──────────────────────────┘
  │ 对得上
  ▼
┌──────────────────────────┐
│ 回写 ~/.opencli/sites/   │  endpoints / field-map / notes / fixtures
└──────────────────────────┘
  │
  ▼
DONE

Runbook(一步一步勾选)

[ ] 1. opencli doctor 返回 "Everything looks good"
[ ] 2. 读站点记忆:
       [ ] ~/.opencli/sites/<site>/endpoints.json 存在?里面有想要的 endpoint?
       [ ] references/site-memory/<site>.md 存在?看"已知 endpoint"节
       [ ] 命中后:**跳到第 5(endpoint 验证) + 第 7(字段核对)**,不能直接跳第 9 写 adapter
       [ ] memory 写入超过 30 天(看 `verified_at`)→ 当作过期,按冷启动走 Step 3 → 4
[ ] 3. 侦察(site-recon.md):
       [ ] **首选**:`opencli browser analyze <url>` 一步拿 pattern + 反爬 + 最近 adapter + next step
       [ ] `analyze` 结论模糊时再手跑:`open` → `wait time 2` (或 `wait xhr <regex>`) → `network`
       [ ] 定 Pattern(A / B / C / D / E)
[ ] 4. API 发现(api-discovery.md)按 Pattern 选 §:
       [ ] Pattern A → §1 network 精读
       [ ] Pattern B → §2 state 抽取 + §1 深层数据
       [ ] Pattern C → §3 bundle / script src 搜索
       [ ] Pattern D → §4 token 来源 + 降级 §5
       [ ] Pattern E → 找 HTTP 轮询接口;找不到才 §5
       [ ] 无文档 API / DOM 丢数据 / 写操作 / bundle 与 network 冲突 → `deep-recon.md`
           [ ] 写 intent matrix 和明确的 mutation boundary
           [ ] baseline → 单一动作 → 新请求 diff;至少一组 changed-input 对照
           [ ] jsluice 只扩大候选面;候选必须进入 evidence ledger
           [ ] read 候选过 occurrence/replay/completeness/auth/pagination/failure gate
           [ ] write 候选有明确授权、目标绑定、幂等/不确定性与不可自动重试语义
[ ] 5. 候选合同验证(memory 命中也要重跑):
       [ ] `PUBLIC_API / COOKIE_API / PAGE_FETCH`:safe replay 跨两个输入返回成功
       [ ] `INTERCEPT`:两次自然页面动作都截到属于目标 identity 的完整响应
       [ ] 响应含目标数据(不是 HTML / 广告 / 推荐侧栏),字段与网页对得上
       [ ] 分页达到 exact limit 或证明 upstream exhaustion;失败不返回 partial
       [ ] write 不自动 replay,必须过 `deep-recon.md` 的额外合同门禁
[ ] 6. 写 strategy note(写代码前的强制产物):
       [ ] 从 `PUBLIC_API / COOKIE_API / PAGE_FETCH / INTERCEPT / DOM_STATE / UI_SELECTOR` 选一个
       [ ] 填 Contract:`stable / visible-ui / internal-unstable`
       [ ] 填 Evidence:observed request/state、auth source、replay result
       [ ] 如果选 `PAGE_FETCH` / `INTERCEPT`,必须解释为什么 `PUBLIC_API` / `COOKIE_API` / `UI_SELECTOR` / `DOM_STATE` 都不适合
       [ ] 如果选 `UI_SELECTOR` / `DOM_STATE`,不需要为 "为什么不是 API" 过度辩护;只要说明语义锚点和 typed error 路径
[ ] 7. 字段解码:
       [ ] 自解释 → 直接用 key
       [ ] 已知代号 → field-conventions.md 查表
       [ ] 未知代号 → field-decode-playbook.md(排序键对比 / 结构差分 / 常量排查)
[ ] 8. 设计 columns(output-design.md):
       [ ] 命名 camelCase 且对齐邻居 adapter
       [ ] 类型 / 单位 / 百分比格式清楚
       [ ] 顺序:识别列 → 业务数字 → metadata
[ ] 9. 写 adapter(adapter-template.md):
       [ ] opencli browser init <site>/<name>
       [ ] 找同站点或同类型最像的 adapter,cp 过来
       [ ] 改 name / URL / 字段映射
[ ] 10. opencli browser verify <site>/<name>
        [ ] 首轮通过后立刻 `--write-fixture` 生成 `~/.opencli/sites/<site>/verify/<cmd>.json` 种子
        [ ] 手改种子:加 `patterns`(URL / 日期 / ID 格式)+ `notEmpty`(核心字段)+ 收紧 `rowCount`
        [ ] 再跑一次 `opencli browser verify <site>/<name>`,确认 ✓ matches fixture
[ ] 11. 字段值 vs 网页肉眼比对(别只看 "Adapter works!")
[ ] 12. 回写站点记忆(**verify 通过 + 肉眼比对对得上之后**,schema 见 `references/site-memory.md`):
        [ ] `endpoints.json`:以 endpoint 的短名为 key,value = `{url, method, params.{required,optional}, response, verified_at: YYYY-MM-DD, notes}`
        [ ] `field-map.json`:只追加新代号。key = 字段代号,value = `{meaning, verified_at: YYYY-MM-DD, source}`;**已存在的 key 不要覆盖**,有冲突先和网页肉眼值对齐再写
        [ ] `notes.md`:顶部追加一段 `## YYYY-MM-DD by <agent/user>`,写本次写 adapter 时遇到的新坑 / 新结论
        [ ] `verify/<cmd>.json`:**必填。** `opencli browser verify` 的期望值(args / rowCount / columns / types / patterns / notEmpty),Step 10 已经让你生成了,这里只是 checklist
        [ ] `fixtures/<cmd>-<YYYYMMDDHHMM>.json`:仅保存公开数据或可证明完成脱敏的样本;私人邮箱/消息/账号等高敏响应改用合成 fixture,不落盘
        [ ] 原始 dump/capture 只短暂落 `/tmp/` 或受控 cache;安全分级后的长期样本才进 `fixtures/`,任务结束清理原始文件
[ ] 13. repo 贡献收口(私人 adapter 可跳过):
        [ ] production-path tests,不只测 parser/helper
        [ ] `npm run typecheck` + focused/site tests + `npm run build`
        [ ] `node dist/src/main.js validate <site>`
        [ ] `npm run check:typed-error-lint` + `npm run check:silent-column-drop`
        [ ] adapter 文档;若 site memory 有稳定新知识则同步
        [ ] `git diff --check` + 敏感数据扫描 + 删除 raw capture/cache + 释放 browser session
        [ ] 写操作或私有协议请独立 review exact head 后再合入

降级路径(某步卡住跳到哪)

卡在现象跳去
Step 4 API 发现network 空,__INITIAL_STATE__ 也空§3 bundle 搜 baseURL
bundle 搜不到 baseURL§5 intercept
Step 5 endpoint 验证401 / 403§4 token 排查
200 但响应是 HTML回 Step 3 换 Pattern 判断
200 但 data: [] 空参数传错 / 接口换版,回 §1 看 network 里真实请求头
Step 7 字段解码排序键对比推不出field-decode-playbook.md §3 结构差分
还推不出先输出 raw,adapter 跑起来再迭代
Step 10 verify 失败fltt 漏了 / 字段映射错autofix skill;复现命令加 --trace retain-on-failure
某列永远是 null字段路径错了,回 Step 7
Step 10 verify fixture mismatch[pattern] row[i] 报错先肉眼比对网页值;值对 → 是 fixture pattern 太严,放宽;值不对 → 字段映射错
[column] missing column "X"实际 response 没这列(站点改版 or args 影响);重新 --update-fixture 或修 adapter
[type] actual null / undefined字段提取失败,回 Step 7 重抽;临时 fallback 用 union type string|null 只有在语义真的可空时用
Step 11 数值不对差 10000 倍单位不统一("万" vs "元")
百分比小 100 倍响应已是 0.025,不要 × 100

Show full SKILL.md (282 more words)Show less

参考文件

文件什么时候翻
references/coverage-matrix.md动手前做"是否在范围内"自测
references/site-recon.mdStep 3 定站点类型
references/api-discovery.mdStep 4 找 endpoint
references/deep-recon.md复杂无文档站:动作归因、jsluice 候选扩展、合同证明、读写安全与交付净账
references/strategy-selection.mdStep 6 填 strategy note 之前:契约模型 + 实测 fix 频率 + api_candidates 证据用法 + 反例
references/field-conventions.mdStep 7 查已知字段代号
references/field-decode-playbook.mdStep 7 字段不在词典时
references/output-design.mdStep 8 命名 / 类型 / 顺序
references/adapter-template.mdStep 9 文件结构 + 活例子 convertible.js
references/site-memory.md总览:in-repo 种子 + 本地 ~/.opencli/sites/ 的两层结构
references/site-memory/<site>.mdStep 2 读站点公共知识(eastmoney / xueqiu / bilibili / tonghuashun 已铺)
references/success-rate-pitfalls.mdStep 7 / 11 踩坑前翻:11 种"verify 能过但数据是错的"静默失败(含 aria-label locale-dependence)
references/jsdom-fixture-pattern.md当 adapter 走 page.evaluate 内 DOM 抽取、且 mocked-evaluate 单测漏 silent bug 时——把 HTML 冻进 clis/<site>/__fixtures__/ 用 JSDOM 跑(含 fixture 创建 mandatory awk 'NF>0' 收紧 + reverse-validate 纪律)
references/typed-errors.md写 func 主体之前必读:5 类 typed error 落点表(ArgumentError / EmptyResultError / CommandExecutionError / AuthRequiredError / TimeoutError)+ 三大 silent anti-pattern(silent-clamp / sentinel-row / generic CliError)的反例修法

关键约定

  • adapter 只引 @jackwener/opencli/registry + @jackwener/opencli/errors,不用第三方
  • columns 数组和 func 返回对象 keys 完全对齐(含顺序)
  • 中间解析对象 key 不能跟 columns 任一项重叠(否则 silent-column-drop audit 误判,PR #1329 R1 真踩过;改成专属命名 + push row 时 destructure aliasing)
  • browser: field 决定 func 签名:browser:false → (args),browser:true → (page, args)。搞反时 args 实际是 debug flag,所有外部参数 silent fallback 到 default(PR #1329 upstream 之前 8 个 non-browser adapter 全踩过这个)
  • 已知失败按 references/typed-errors.md 5-classification 抛对应 typed error;不要 silent return [],不要 silent return [{sentinel}],不要 Math.max/min silent clamp 外部参数
  • 写私人 adapter 用 ~/.opencli/clis/<site>/<name>.js(免 build);要提 PR 才 copy 到 clis/<site>/<name>.js
  • 站点记忆每轮回写:没记忆 → 用 skill → 产生记忆 → 下次变 5 分钟
  • “真实发生过”不等于“可作为 production contract 重放”。私有写请求、一次性风控 token、页面 runtime controller 都必须过 deep-recon.md 的 contract gate;过不了就记录 blocker/lift condition,不生成伪 API 命令。
  • 调试过程中的原始 dump / 抓包 / HTML 样本只能短暂落在系统 /tmp/ 或受控 cache,任务结束删除。只有通过 site-memory.md 数据分级、准备长期保留的公开/合成/已脱敏样本才进入 ~/.opencli/sites/<site>/fixtures/。严禁在 repo 根目录、clis/<site>/ 或当前工作目录留 .dbg-*.html / raw-*.json / sample.*。
  • JSDOM unit-test fixture(clis/<site>/__fixtures__/<command>.html)是上面那条的例外——它是有意 commit 进 repo 的 review artifact,不是临时 dump。但因此 quality bar 要更高:必须按 references/jsdom-fixture-pattern.md 的 5 步做完(含 mandatory awk 'NF>0' 空白行收紧),并 reverse-validate 一道证明 regression guard 真能挂。

卡住了

  • 诊断类:opencli doctor → 看 notes.md → 搜 autofix skill
  • 字段解码类:field-decode-playbook.md 全三节走完 → 先输出 raw 迭代
  • endpoint 找不到:api-discovery §5 intercept 兜底

不要猜。猜错了 verify 能通过但数据是错的,用户看到乱码才发现。

© jackwener, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 18 other files (references) in skills/opencli-adapter-author of jackwener/OpenCLI.

  • SKILL.md
  • references/adapter-template.md
  • references/api-discovery.md
  • references/coverage-matrix.md
  • references/deep-recon.md
  • references/field-conventions.md
  • references/field-decode-playbook.md
  • references/jsdom-fixture-pattern.md
  • references/output-design.md
  • references/site-memory.md
  • references/site-memory/bilibili.md
  • references/site-memory/eastmoney.md
  • references/site-memory/gmail.md
  • references/site-memory/tonghuashun.md
  • references/site-memory/xueqiu.md
  • references/site-recon.md
  • references/strategy-selection.md
  • references/success-rate-pitfalls.md
  • references/typed-errors.md

Open the folder on GitHubat commit 2413694

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in jackwener/OpenCLI, which our catalogue first saw on October 7, 2026.

Compare with similar skills

OpenCLI Adapter Author next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

OpenCLI Adapter Author compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
OpenCLI Adapter Author this skilljackwener/OpenCLI30k1 repos~3.4kAutomated safety check: PassApache-2.0
Firecrawl Interact Integrationfirecrawl/firecrawl190k1 repos~731Automated safety check: PassISC
Derive API Client from Browser Trafficvercel-labs/agent-browser44k2 repos~1.3kAutomated safety check: PassApache-2.0
Kimi WebbridgeMoonshotAI/kimi-code7.8k—~3.6kAutomated safety check: PassMIT
Kimi Webbridgemxyhi/ok-skills493—~4.3kAutomated safety check: PassApache-2.0
Open Web Bridge Browser Controlwoniu9524/open-web-bridge121—~8.1kAutomated safety check: PassMIT

Similar skills

  • Guides adding Firecrawl's /interact endpoint to product code for pages that need clicks, forms, pagination or logged-in flows beyond plain scraping.

    190k GitHub starsUsed in 1 repo~731 tokens
    Data & AnalyticsAuto-check passed
  • Official

    Records a site's browser traffic into a HAR file, then builds a standalone client or CLI that calls its internal endpoints directly with no browser.

    44k GitHub starsUsed in 2 repos~1.3k tokens
    DevelopmentAuto-check passed
  • Kimi Webbridge

    MoonshotAI/kimi-code

    Kimi Browser Extension(Kimi 浏览器扩展,原 Kimi WebBridge)lets AI control the user's real browser — navigate, click, type, read, screenshot, and interact with any website using the user's actual login…

    7.8k GitHub stars~3.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Kimi Webbridge

    mxyhi/ok-skills

    Kimi Browser Extension(Kimi 浏览器扩展,原 Kimi WebBridge)lets AI control the user's real browser — navigate, click, type, read, screenshot, and interact with any website using the user's actual login…

    493 GitHub stars~4.3k tokensUpdated 7 days ago
    DevelopmentAuto-check passed
  • Open Web Bridge Browser Control

    woniu9524/open-web-bridge

    Lets the agent operate your own logged-in browser through the owb command to read pages, fill forms, audit sites and capture network traffic.

    121 GitHub stars~8.1k tokensUpdated 1 mo ago
    Productivity & AutomationAuto-check passed
  • Dev Browser Automation

    MemTensor/MemOS

    Automates a real browser through short TypeScript scripts that keep page state between runs, for navigating, filling forms, taking screenshots and extracting data.

    12k GitHub starsUsed in 3 repos~1.7k tokens
    Productivity & AutomationAuto-check passed

More from jackwener/OpenCLI

  • Routes a search question to the most suitable opencli source, normally one AI site plus up to two specialist sites, with call limits per question and a recap at the end.

    30k GitHub starsUsed in 2 repos~753 tokens
    Auto-check passed
  • OpenCLI Browser Control

    jackwener/OpenCLI

    Drives a real Chrome window through the opencli CLI to inspect pages, fill forms, click through logged-in flows and extract data from structured command responses.

    30k GitHub starsUsed in 2 repos~7.3k tokens
    Auto-check passed
  • OpenCLI Adapter Autofix

    jackwener/OpenCLI

    Repairs a broken OpenCLI site adapter after a command fails: collects a trace, patches only the adapter, retries, and files an upstream GitHub issue once fixed.

    30k GitHub starsUsed in 1 repo~3.2k tokens
    Auto-check passed
  • OpenCLI Usage Map

    jackwener/OpenCLI

    Orients an agent on OpenCLI, which turns websites and Electron apps into opencli site-command calls, covering adapters, strategies, install and which skill to load next.

    30k GitHub starsUsed in 1 repo~2.3k tokens
    Auto-check passed

Questions about OpenCLI Adapter Author

What does OpenCLI Adapter Author do?

Walks through writing an OpenCLI adapter for a new site or a new command on an existing one, from first recon and field decoding to coding and verification. The goal is for a simple site to go from nothing to a passing `opencli browser verify` within about 30 minutes, while complex, private-protocol or write-operation sites put complete evidence and safety ahead of the time limit. The agent uses only existing commands (`opencli browser`, `opencli doctor`, `opencli browser init`, `opencli browser verify`) and debugs browser adapters with tracing on and the tab kept in the foreground.

When should I use OpenCLI Adapter Author?

OpenCLI Adapter Author fits situations like: writing an OpenCLI adapter for a new website; adding another command to an existing site adapter; choosing between a public API, cookie API and page scraping for a site.

How do I install OpenCLI Adapter Author in Claude Code?

Run `npx skills add jackwener/OpenCLI --skill opencli-adapter-author -a claude-code`. Or copy the skill folder (skills/opencli-adapter-author in jackwener/OpenCLI) into .claude/skills/opencli-adapter-author in your project. Claude Code loads it when a task matches its description.

How do I install OpenCLI Adapter Author in Codex?

Run `npx skills add jackwener/OpenCLI --skill opencli-adapter-author -a codex`. Or copy the skill folder (skills/opencli-adapter-author in jackwener/OpenCLI) into .agents/skills/opencli-adapter-author in your project. Codex loads it when a task matches its description.

Can I use OpenCLI Adapter Author in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jackwener/OpenCLI --skill opencli-adapter-author -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/opencli-adapter-author, .gemini/skills/opencli-adapter-author, .github/skills/opencli-adapter-author and .opencode/skills/opencli-adapter-author in your project.

What does OpenCLI Adapter Author need to run?

SKILL.md names no scripts, command-line tools or credentials: OpenCLI Adapter Author is instructions for the agent only. Our summary lists: The `opencli` CLI with a logged-in browser session; `opencli doctor` reporting a healthy setup. Its frontmatter pre-approves these tools: Bash(opencli:*), Bash(jsluice:*), Read, Edit, Write, Grep.

Does OpenCLI Adapter Author access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is OpenCLI Adapter Author safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does OpenCLI Adapter Author use?

OpenCLI Adapter Author is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does OpenCLI Adapter Author use?

About 3.4k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 36k tokens, read only when the agent opens those files.

What are the alternatives to OpenCLI Adapter Author?

Skills that share tags, products or a category with OpenCLI Adapter Author: Firecrawl Interact Integration (firecrawl/firecrawl, 190k stars), Derive API Client from Browser Traffic (vercel-labs/agent-browser, 44k stars), Kimi Webbridge (MoonshotAI/kimi-code, 7.8k stars) and Kimi Webbridge (mxyhi/ok-skills, 493 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains OpenCLI Adapter Author?

jackwener (a GitHub user) maintains it in jackwener/OpenCLI, which has 29,917 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on September 24, 2026.

Source: jackwener/OpenCLI on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.