跳到正文
cccskills
免费在 GitHub 公开

api-state

Fetch and report current API syntax, changelog entries, and breaking changes for a specific library or protocol version. One research angle within a parallel technical-research set — runs independently and returns a structured cited report. Invoke when a specific library name and version are the target.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.2 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

API State

One research angle in a multi-angle technical research system. Singular focus: current API syntax, changelog, and breaking changes for a specific library or protocol version.

Runs independently. No shared state with other research angles during execution. Return all findings as content — do not write to any backlog item.

Workflow

Step 1 — Scope

Research target received from orchestrator:

$ARGUMENTS

Confirm the technology name, version, and specific API surface before proceeding. If any are missing or ambiguous, state the gap and output:

STATUS: BLOCKED Reason: [missing field] — cannot proceed without [technology name | version | API surface].

Step 2 — Current API State

Fetch the primary documentation page for the technology. Use mcp__Ref__ref_search_documentation to locate the canonical docs URL, then mcp__Ref__ref_read_url to retrieve the page. If the library is not indexed by Ref, use context7 (mcp__context7__resolve-library-id then mcp__context7__query-docs). Derive the URL from the confirmed scope — do not rely on training data for the URL or for version-specific syntax.

Extract and record:

  • Exact method/class/decorator signatures
  • All parameter forms (required, optional, typed, variadic)
  • Return types and async behaviour
  • Context injection patterns (if applicable to the technology)
  • Any configuration or initialization requirements

Cite each item with the source URL and access date.

Step 3 — Changelog

Fetch the raw changelog. Locate it at the canonical path for the technology — commonly CHANGELOG.md or CHANGES.md in the GitHub repository, or the GitHub Releases page. Use mcp__Ref__ref_read_url for the changelog URL. Fall back to WebFetch for raw GitHub URLs not served by Ref.

Extract all entries relevant to the API surface confirmed in Step 1, version by version. Cite each entry with its version number and source URL.

If the changelog is absent or the relevant entries are not present, state this explicitly: "Changelog not found at [URLs attempted]" or "No changelog entries for [API surface] found in versions [range searched]."

Step 4 — Breaking Changes

From the changelog and release notes, identify changes between the most recent major versions:

  • What broke silently (no deprecation warning before removal)
  • What requires explicit migration (changed signatures, removed parameters, renamed classes)
  • What was deprecated and when (deprecation version and removal version if stated)

If no breaking changes are found, state this explicitly with the version range examined.

Step 5 — Gotchas

Synthesise from the findings in Steps 2–4:

  • What was the changelog FIXING? (implies the behaviour was broken before that version)
  • What do the docs explicitly warn about?
  • What behaviour is client-defined versus protocol-defined or library-defined?

Every gotcha must cite its source: a changelog entry URL, a doc warning with URL, or an observed inconsistency between two cited sources.

Step 6 — Output

Return the following structured report, populated from the findings above, then close with a STATUS block:

## API State — {technology} {version} — {date}

### Current API
[exact syntax, parameters, return types — with cited source URL and access date]

### Changelog (relevant entries)
[version: entry — source URL]

### Breaking Changes ({prev version} → {current version})
[each change with source]

### Gotchas
[each item with source — changelog entry, doc warning, or observed evidence]

### Gaps
[what was searched for but not found — explicit list]

STATUS: DONE Findings delivered to caller above. No backlog item written.

The Gaps section is mandatory. If nothing is missing, write: "No gaps — all expected information was found in primary sources."

Constraints

  • "Not documented in [source]" is a valid finding — state it rather than omitting it.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月20日 更新

查看 Jamie-BitFlight 的全部技能

报告此技能的问题