跳到正文
cccskills
免费在 GitHub 公开

ase-help-skill

Show the manual page of an ASE skill, addressed by its full name, by any abbreviation of it, or by a description of its purpose, and list the entire skill catalog when no name is given. Use when the user wants the "manual", "man page", "manpage", or "help" of a particular ASE skill, or asks what a certain `ase-xxx-xxx` skill does.

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(3)

  • SKILL.md7.5 KB
  • catalog.md3.3 KB
  • help.md2.6 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

@${CLAUDE_SKILL_DIR}/../../meta/ase-control.md @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md @${CLAUDE_SKILL_DIR}/../../meta/ase-dialog.md @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md

<purpose name="ase-help-skill"> Show the Manual Page of an ASE Skill </purpose>

<expand name="getopt" arg1="ase-help-skill" arg2=""> $ARGUMENTS </expand>

<objective> *Show* the *manual page* of the ASE skill addressed by the following skill name, abbreviation of it, or description of its purpose: <skill-ref><getopt-arguments/></skill-ref> </objective>

The following <catalog/> is the index of all ASE skills -- one ⎈ **<group/>** entry per skill group and one ○ `<name/>`: <purpose/> entry per skill -- and this is the sole index <skill-ref/> is resolved against:

<catalog> @${CLAUDE_SKILL_DIR}/catalog.md </catalog> <flow>
  1. <step id="STEP 1: Resolve Skill Name">
    1. <if condition="<skill-ref/> is empty"> No particular skill was addressed, so render the *entire* <catalog/> as a browsable list with the following <template/> -- one list entry per catalog entry, in catalog order, where <name/> and <purpose/> are the two fields of the entry (and <name-padded/> is <name/>, padded to 22 characters with spaces on the right) -- and then immediately *STOP* processing the entire current skill: <template> <ase-tpl-head title="SKILL CATALOG"/> <catalog/> <ase-tpl-foot title="SKILL CATALOG"/> </template>

      The rendered <catalog/> itself was explicitly requested and hence is always emitted. Only the trailing pointer to the per-skill manual page is a hint, so emit it by expanding the following (which, depending on the configured <ase-guidance-level/>, may expand into nothing):

      <ase-tpl-hint level="normal"> run `/ase-help-skill ase-xxx-xxx` for manual page of individual skill </ase-tpl-hint> </if>
    2. Set <skill-ref-raw/> to <skill-ref/> with only its leading and trailing whitespace stripped, as the verbatim wording of the user is required later on.

      Normalize <skill-ref/> by stripping all leading and trailing whitespace and then, repeatedly, any leading / and ase: prefix, so that ase-code-lint, /ase-code-lint, ase:ase-code-lint, and /ase:ase-code-lint all normalize to ase-code-lint. Do not output anything.

    3. Resolve the normalized <skill-ref/> against <catalog/> in three tiers and store the outcome in <candidates/>. Each tier is tried only if all preceding tiers yielded no candidate at all:

      1. Exact Name Tier:

        If a catalog name is equal to <skill-ref/>, set <candidates/> to exactly that single name.

      2. Substring Name Tier:

        Set <candidates/> to all catalog names containing <skill-ref/> as a substring, in alphabetical order.

      3. Fuzzy Purpose Tier:

        Set <candidates/> to all catalog names whose purpose -- the part after the colon of the catalog entry -- fuzzily matches <skill-ref-raw/>, in descending order of match quality. Match against <skill-ref-raw/>, and not against <skill-ref/>, as this tier matches free-text wording, which the normalization of sub-step 2 would distort.

        A purpose matches fuzzily if it shares the topic, the wording, or evident synonyms with <skill-ref-raw/>, so that e.g. manpage matches Show the Manual Page of an ASE Skill and root cause matches Five-Whys Root-Cause Analysis. Include plausible matches only -- if none is plausible, leave <candidates/> empty.

      Set <count/> to the number of entries in <candidates/>. Do not output anything.

    </step>
  2. <step id="STEP 2: Dispatch Resolution">
    1. <if condition="<count/> is equal 0"> Only output the following <template/> and then immediately *STOP* processing the entire current skill: <template> ⧉ **ASE**: ✪ skill: **ase-help-skill**, ▶ ERROR: unknown skill: **<skill-ref-raw/>** </template> </if>
    2. <elseif condition="<count/> is equal 1"> Set <name/> to the single entry of <candidates/> and continue processing. Do not output anything. </elseif>
    3. <else> The abbreviation is *ambiguous*, so let the user pick the intended skill.

      Set <shown/> to the first 9 entries of <candidates/>, as the dialog renders at most nine answer lines.

      <if condition="<count/> is greater than 9"> Set <truncation> (showing 9 of <count/> candidates)</truncation> </if> <else> Set <truncation></truncation> (set to empty) </else>

      In the following, you MUST NOT use your built-in <user-dialog-tool/> tool! Instead, you MUST just show a custom dialog according to the expanded custom-dialog definition. You MUST closely follow this definition.

      Let the user select the intended skill by raising a question with the following custom dialog, where each answer line corresponds to one entry of <shown/>, using the catalog name as the label and its catalog purpose as the description:

      <expand name="custom-dialog" arg1="--no-other"> Ambiguous Skill: Which skill's manual page should be shown?<truncation/> <name/>: <purpose/> [...] </expand>

      Check the <result/> and dispatch accordingly:

      • If <result/> is CANCEL: STOP processing without any further output.

      • Otherwise: Set <name/> to the selected skill name and continue processing.

        </else>
    </step>
  3. <step id="STEP 3: Render Manual Page">
    1. Use the Read tool to read the manual page of the resolved skill <name/> and set <manual/> to its content. The file path is formed by joining <ase-plugin-root/> and skills/<name/>/help.md with exactly one / separator, as <ase-plugin-root/> may or may not carry a trailing /. Do not output anything related to this tool call.

    2. <if condition="<manual/> is empty or could not be read"> Only output the following <template/> and then immediately *STOP* processing the entire current skill: <template> ⧉ **ASE**: ✪ skill: **ase-help-skill**, ▶ ERROR: unreadable manual page: **<name/>** </template> </if>
    3. Treat <manual/> as verbatim Markdown. You MUST NOT truncate, summarize, reformat, or partially show it. Only output the following <template/>:

      <template> <ase-tpl-head title="MANUAL PAGE: <name/>"/> <manual/> <ase-tpl-foot title="MANUAL PAGE: <name/>"/> </template>
    </step>
</flow>

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

Review software architecture, including package cohesion and inter-package coupling

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

Discover additional, third-party components (libraries/frameworks) for the technology stack to provide needed functionality.

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

Analyze the source code for problems in either the logic and semantics and its related control flow, performance and efficiency, or security.

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

Edit Source Code: Use when the user wants to "edit" the code base in one shot from a query or a bare analyzer issue id like "P1", fusing crafting, refactoring, and resolving with optional grilling, verification, looping, and Git worktree isolation.

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

Explains code with WHAT, WHY, ANALOGY, DIAGRAM, CRUXES, and GOTCHAS. Use when you want to know how code works or when the user asks "how does this work?"

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

Give insights into the source code.

中文简介正在准备中。当前显示原文说明。

rse/ase602026年10月22日 更新

查看 rse 的全部技能

报告此技能的问题