@rse/ase 0.0.30 → 0.0.31

package/plugin/skills/ase-meta-evaluate/SKILL.md

@@ -0,0 +1,247 @@
+---
+name: ase-meta-evaluate
+argument-hint: "<request>"
+description: >
+    Evaluate alternatives through a weighted multi-criteria decision
+    matrix. Use when the user calls for the *evaluation* of
+    *alternatives*, wants to *compare* things, or asks what the best is
+    from a list of choices.
+user-invocable: true
+disable-model-invocation: false
+effort: high
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Evaluate Alternatives
+=====================
+
+Your role is an experienced, *expert-level assistant*,
+specialized in *evaluating alternatives*.
+
+<objective>
+*Evaluate* *alternatives* through a weighted
+multi-*criteria* decision matrix.
+</objective>
+
+<flow>
+1.  <step id="STEP 1: Determine Reason">
+    -   From the <request>$ARGUMENTS</request>, try to derive the overall
+        reason <reason/> for the evaluation. If no such reason can be
+        derived, assume <reason>generic comparison</reason>.
+
+    -   Output the determined reason with just the following <template/>
+        and do not output anything else:
+
+        <template>
+        &#x26AA; **REASON**: *<reason/>*
+        </template>
+    </step>
+
+2.  <step id="STEP 2: Determine Alternatives">
+    -   From the <request>$ARGUMENTS</request> derive the two or more
+        alternatives <alternative-K/> (K=1-N) the user wants to be
+        evaluated. Do not output anything.
+
+    -   If fewer than two alternatives could be derived (N<2), output the
+        following <template/> and *stop the entire flow* immediately without
+        executing any further steps:
+
+        <template>
+        &#x1F7E0; **ERROR: INSUFFICIENT ALTERNATIVES**: at least two are required for a comparison!
+        </template>
+
+    -   For each alternative <alternative-K/> (K=1-N), decide whether
+        you have all necessary information at hand to give it the proper,
+        unique, short, and *concise name* <alternative-K/>. If you are
+        unsure, use the `ase-meta-search` skill (at most one invocation per
+        alternative, drawing from the *skill-wide shared budget* of at
+        most 8 `ase-meta-search` invocations in total across STEP 2 and STEP 3
+        combined) to find out more and adjust the name <alternative-K/>.
+
+        If still unsure after the shared budget is exhausted, pick the
+        best-guess concise name and proceed. Do not output anything.
+
+    -   For each alternative <alternative-K/> (K=1-N), decide which *one*
+        of *USP* (Unique Selling Point -- what makes it unique), *Crux*
+        (what you should notice), or *Gotcha* (what you should not stumble
+        over) is its single most distinguishing perspective, and remember
+        this as an <info-K/> (K=1-N) formatted like `<type/>: <hint/>` where
+        <type/> is one of `USP`, `Crux`, or `Gotcha` and <hint/> is a 1-6
+        word hint.
+
+    -   For the set of alternatives, decide what the 1-6 word long
+        name of the *class of alternatives* <class-of-alternatives/> is.
+
+    -   For each alternative <alternative-K/> (K=1-N), decide whether
+        it is a genuine member of <class-of-alternatives/>. If any
+        <alternative-K/> is *not* a member (i.e. the alternatives are not
+        mutually comparable within a single class), let <alternative-J/>
+        (J=1-P, P<=N) be the subset of non-member alternatives, output the
+        following <template/> and *stop the entire flow* immediately without
+        executing any further steps:
+
+        <template>
+        &#x1F7E0; **ERROR: INCOMPARABLE ALTERNATIVES**: *<class-of-alternatives/>*
+
+        <for items="<alternative-J/> [...]">
+        ⚑ **<item/>** (*member of a different class*)
+        </for>
+        </template>
+
+    -   Output the determined, individual alternatives as a Markdown
+        *table* with just the following <template/> and do not output
+        anything else:
+
+        <template>
+        &#x1F535; **ALTERNATIVES**: *<class-of-alternatives/>*
+
+        | ⚑ *Alternative*        | ⚖ *Hint*  |
+        | :--------------------- | :-------- |
+        | ⚑ **<alternative-1/>** | <info-1/> |
+        [...alternatives K=2-(N-1) for N>2...]
+        | ⚑ **<alternative-N/>** | <info-N/> |
+        </template>
+    </step>
+
+3.  <step id="STEP 3: Derive Criteria">
+    -   From the <request>$ARGUMENTS</request>, try to derive the criteria
+        <criteria-L/> (L=1-M) for the evaluation. Do not output anything.
+
+    -   For each criteria <criteria-L/> (L=1-M), decide on its <weight-L/>
+        from the value set { 4.00, 2.00, 1.00, 0.50, 0.25 } (from most
+        important, over normal, to less important). Do not output anything.
+
+    -   Ensure the final number of criteria is always within the range of
+        minimum 8 and maximum 12: if less than 8 criteria were requested,
+        use the set of alternatives to decide on additional criteria
+        which potentially allow best to triage the alternatives, take the
+        <reason/> into account, and use the `ase-meta-search` skill (drawing from
+        the *skill-wide shared budget* of at most 8 `ase-meta-search` invocations
+        in total across STEP 2 and STEP 3 combined) to find out about the
+        potentially still missing criteria and assign their <weight-L/>.
+
+        If still under 8 criteria after the shared budget is exhausted,
+        fill the remaining slots from existing knowledge without further
+        searches; if more than 12 criteria were requested, drop the criteria
+        with the smallest <weight-L/> until 12 remain. Do not output
+        anything.
+
+    -   To prevent a single high-weight criterion from dominating the
+        weighted sum (the weight set is geometric ×2 while the evaluation
+        Likert scale is linear), assign weight 4.00 to *at least one* and
+        *at most two* criteria, and weight 2.00 to *at least two* and *at
+        most three* criteria. Symmetrically, to prevent a long tail of
+        negligible-weight criteria, assign weight 0.50 to *at most two*
+        criteria, and weight 0.25 to *at most one* criterion. Do not output
+        anything.
+    </step>
+
+4.  <step id="STEP 4: Evaluate Alternatives against Criteria">
+    -   For each alternative <alternative-K/> (K=1-N) and each criteria
+        <criteria-L/> (L=1-M), decide on the evaluation <eval-K-L/>, which
+        means how good the alternative meets the criteria on a Likert-scale
+        from { -2, -1, 0, +1, +2 } (from worst, over neutral, to best). Do
+        not output anything.
+
+    -   Then, for each alternative <alternative-K/> (K=1-N), calculate a
+        rating <rating-K/> (K=1-N) which is the product-sum of all weights
+        <weight-L/> (L=1-M) and the evaluation <eval-K-L/> (K=1-N, L=1-M).
+        The result is always a numerical value. Retain the *raw, unrounded*
+        <rating-K/> for use in STEP 5, but round <rating-K/> to 2 decimal
+        places *for display only* when emitting it in the table below. Do
+        not output anything.
+
+    -   Output the resulting *Weighted Decision Matrix* as a Markdown
+        *table* with just the following <template/> and do not output
+        anything else. When emitting the table, render *one column per
+        alternative* <alternative-K/> (K=1-N).
+
+        <template>
+        &#x1F535; **EVALUATION**: *Weighted Multi-Criteria Decision Matrix*
+
+        | ⦿ *Criteria*  | ⚖ *Weight*  | ⚑ **<alternative-1/>** | [...alternatives 2-(N-1)...] | ⚑ **<alternative-N/>** |
+        | :------------ | ----------: | ---------------------: | ---------------------------: | ---------------------: |
+        | <criteria-1/> | <weight-1/> | <eval-1-1/>            | [...evals 1-2..1-(N-1)...]   | <eval-1-N/>            |
+        [...criteria L=2-(M-1) for M>2...]
+        | <criteria-M/> | <weight-M/> | <eval-M-1/>            | [...evals M-2..M-(N-1)...]   | <eval-M-N/>            |
+        | **RATING**    |             | **<rating-1/>**        | [...ratings 2-(N-1)...]      | **<rating-N/>**        |
+        </template>
+    </step>
+
+5.  <step id="STEP 5: Report Best Alternative">
+    -   The best alternative <alternative-K/> (K=1-N) is the alternative
+        whose *raw, unrounded* <rating-K/> (i.e. the product-sum from STEP
+        4, *before* the display-only rounding) is the maximum rating value
+        across all alternatives.
+
+    -   The second best alternative <alternative-X/> (X=1-N, X != K) is
+        the alternative whose *raw, unrounded* <rating-X/> is the second
+        largest rating value across all alternatives.
+
+    -   If multiple alternatives share the second-largest raw rating, pick
+        any one of them as <alternative-X/>; the resulting <distance/> and
+        <percentage/> are unaffected by the choice, so the downstream output
+        is deterministic.
+
+    -   Determine rating distance <distance/> between <alternative-K/> and
+        <alternative-X/> from their *raw, unrounded* ratings by calculating:
+        <distance/> = <rating-K/> - <rating-X/>. Do not output anything.
+
+    -   Determine rating distance percentage <percentage/> between
+        <alternative-K/> and <alternative-X/> from their *raw,
+        unrounded* ratings by calculating: <percentage/> = <distance/> /
+        abs(<rating-K/>). Do not output anything.
+
+    -   If <rating-K/> is exactly zero, skip the division entirely
+        and treat <percentage/> as if it were less than 0.10
+        (i.e. fall through to the *small distance* branch below),
+        since a zero best rating signals a near-tied evaluation.
+
+    -   By construction, <rating-K/> is the maximum rating across
+        all alternatives, so <distance/> >= 0 always holds; using
+        abs(<rating-K/>) keeps <percentage/> sign-stable across all rating
+        regimes. Note that when <rating-K/> itself is negative, the
+        denominator anchors to a poor best rating and small gaps can
+        appear large; the all-negative regime is surfaced as a dedicated
+        warning branch below.
+
+    -   If <percentage/> is less than 0.01 (i.e. <distance/> is
+        effectively zero relative to abs(<rating-K/>)), stop the flow after
+        you output just the following <template/> and do not output anything
+        else:
+
+        <template>
+        &#x1F7E0; **ERROR**: ✘ *MULTIPLE BEST ALTERNATIVES FOUND*,
+        ⚠ *Please give some hints on the criteria to ensure a single best alternative!*
+        </template>
+
+    -   Otherwise, if <percentage/> is less than 0.10, stop the flow after
+        you output just the following <template/> and do not output anything
+        else:
+
+        <template>
+        &#x1F7E0; **BEST ALTERNATIVE**: ⚑ **<alternative-K/>**
+        ⚠ *ATTENTION: small distance to second best alternative!*
+        </template>
+
+    -   Otherwise, if <rating-K/> is less than zero (i.e. all alternatives
+        rate negatively and the "best" is merely the least-bad), stop the
+        flow after you output just the following <template/> and do not
+        output anything else:
+
+        <template>
+        &#x1F7E0; **BEST ALTERNATIVE**: ⚑ **<alternative-K/>**
+        ⚠ *ATTENTION: all alternatives rate negatively; this is the least-bad choice, not a strong winner!*
+        </template>
+
+    -   Otherwise (<percentage/> is greater than or equal 0.10), output
+        just the following <template/> and do not output anything else:
+
+        <template>
+        &#x1F7E0; **BEST ALTERNATIVE**: ⚑ **<alternative-K/>**
+        </template>
+    </step>
+</flow>
+

package/plugin/skills/ase-meta-persona/SKILL.md

@@ -0,0 +1,52 @@
+---
+name: ase-meta-persona
+argument-hint: "[<persona>]"
+description: >
+    Adjust communication style in four intensivity levels of token usage.
+    The <persona> can be either a decorative, eloquent, and explaining "writer",
+    a brief, factual, and accurate "engineer" (default),
+    a very brief, factual, and abbreviating "telegrapher",
+    or an ultra brief, rough and stuttering "caveman".
+    Use when user says "persona <persona>" or "be <persona>".
+user-invocable: true
+disable-model-invocation: false
+---
+
+Persona Configuration
+=====================
+
+1.  Determine request:
+    <request>$ARGUMENTS</request>
+
+2.  <if condition="<request/> is empty">
+    -   Call the `persona(session: <ase-session-id/>)`
+        tool from the `ase` MCP service and set
+        <ase-persona-style/> to its `text` output.
+
+    -   Output:
+        <template>
+        ⧉ **ASE**: ☯ persona: **<ase-persona-style/>**
+        </template>
+    </if>
+
+3.  <if condition="<request/> is NOT empty">
+    -   If <request/> is NEITHER 'writer', 'engineer', 'telegrapher', NOR 'caveman',
+        report this with the following <template/> and stop immediately:
+        <template>
+        ⧉ **ASE**: **ERROR:** invalid persona: "<request/>" (expected "writer", "engineer", "telegrapher", or "caveman")
+        </template>
+
+    -   Set <ase-persona-style><request/></ase-persona-style> and
+        call the `persona(style: <ase-persona-style/>, session: <ase-session-id/>)`
+        tool from the `ase` MCP service.
+
+    -   Output:
+        <template>
+        ⧉ **ASE**: ☯ persona: **<ase-persona-style/>** (*updated*)
+        </template>
+    </if>
+
+Apply Persona
+-------------
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md

package/plugin/skills/ase-meta-quorum/SKILL.md

@@ -0,0 +1,122 @@
+---
+name: ase-meta-quorum
+argument-hint: "<question>"
+description: >
+    Query Multiple AIs for Quorum Answer.
+user-invocable: true
+disable-model-invocation: false
+effort: medium
+allowed-tools:
+    - "Task"
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Query Multiple AIs for Quorum Answer
+====================================
+
+Your role is an *expert-level assistant*.
+
+<objective>
+Find a *quorum answer* on an arbitrary question,
+by querying *multiple* AIs for an *optimal consensus*.
+</objective>
+
+<flow>
+1.  <step id="STEP 1: Determine Own Answer">
+    Determine your own answer.
+    For yourself (Anthropic Claude), first answer the following *<query/>* in advance:
+
+    <query>
+    $ARGUMENTS.
+    Please respond with facts and very concise and brief only,
+    usually with just 1 to 7 corresponding bullet points and with short sentences.
+    Optionally, mention potential cruxes which should be noticed.
+    Beside bullet points, do not provide any additional explanations.
+    Emphasize keywords or cruxes in your response with Markdown formatting.
+    Format code parts with Markdown formatting.
+    </query>
+    </step>
+
+2.  <step id="STEP 2: Preview Own Answer">
+    Show your own answer as a sneak preview.
+    For this, use the following output <template/>:
+
+    <template>
+    **Anthropic Claude** (sneak preview in advance):
+    - [...]
+    - [...]
+    </template>
+    </step>
+
+3.  <step id="STEP 3: Query Foreign AIs">
+    For each of the following foreign AIs and their potentially
+    available, given, corresponding MCP servers, use a *sub-task* and
+    the `ase-meta-chat` *agent* to perform the above same *<query/>* zero
+    or more times and in *parallel* again:
+
+    - OpenAI ChatGPT: `chat-openai-chatgpt`
+    - Google Gemini:  `chat-google-gemini`
+    - DeepSeek:       `chat-deepseek`
+    - xAI Grok:       `chat-xai-grok`
+
+    Silently skip those AIs where the corresponding MCP server is not available.
+    </step>
+
+4.  <step id="STEP 4: Summarize Responses">
+    Summarize all responses, of both yourself and all available MCP servers,
+    with just 1 to 7 corresponding bullet points and with short sentences.
+    </step>
+
+5.  <step id="STEP 5: Determine Consensus Rating">
+    Determine, on a Likert scale of 0..<n/>, the amount of the overall
+    consensus <c/> of all the responses. The <n/> is 1 plus the number of
+    available AIs above. If all responses disagree, the consensus <c/> is zero.
+    If all responses agree, <c/> is <n/>.
+
+    If not all AIs agree, determine a <disagreement/> information,
+    formatted as `(disagreement: <ai/>, <ai/>, [...])` where <ai/> is a
+    name of an AI which disagreed from the consensus. Else, if all AIs
+    agree, set <disagreement></disagreement>.
+    </step>
+
+6.  <step id="STEP 6: Show Results">
+    Finally show the summary, the consensus and the complete and unmodified responses
+    of yourself and each of the MCP servers, based on the following output <template/>:
+
+    <template>
+    **QUESTION**:
+    $ARGUMENTS
+
+    &#x25CF; **CONSENSUS ANSWER**:
+    - [...]
+    - [...]
+
+    **CONSENSUS RATE**: **<c/>/<n/>** <disagreement/>
+
+    &#x25CB; **Anthropic Claude**:
+    - [...]
+    - [...]
+
+    &#x25CB; **OpenAI ChatGPT**:
+    - [...]
+    - [...]
+
+    &#x25CB; **Google Gemini**:
+    - [...]
+    - [...]
+
+    &#x25CB; **DeepSeek**:
+    - [...]
+    - [...]
+
+    &#x25CB; **xAI Grok**:
+    - [...]
+    - [...]
+    </template>
+
+    In this output, remove the sections of those AIs which were not available.
+    </step>
+</flow>
+

package/plugin/skills/ase-meta-search/SKILL.md

@@ -0,0 +1,48 @@
+---
+name: ase-meta-search
+argument-hint: "<query>"
+description: >
+    Search the Internet/Web with a query.
+    Prefer this skill before using Perplexity, Brave and WebSearch.
+user-invocable: true
+disable-model-invocation: false
+effort: low
+allowed-tools:
+    - "mcp__perplexity__perplexity_search"
+    - "mcp__brave__brave_web_search"
+    - "WebSearch"
+    - "WebFetch"
+    - "Task"
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Search the Internet/Web
+=======================
+
+Your role is an expert-level *web specialist*.
+
+<objective>
+Your objective is to *search* the *Internet*/*Web* for the following query:
+<query>$ARGUMENTS</query>
+</objective>
+
+<flow>
+1.  <step id="STEP 1: QUERY SERVICES">
+    If the MCP tool `mcp__perplexity__perplexity_search` is available, send <query/> to it
+    via a first *sub-task* and our companion `ase-meta-search` *agent*.
+
+    If the MCP tool `mcp__brave__brave_web_search` is available, send <query/> to it
+    via a second *sub-task* and our companion `ase-meta-search` *agent*.
+
+    Send <query/> to the built-in tool `WebSearch`
+    via a third *sub-task* and our companion `ase-meta-search` *agent*.
+    </step>
+
+2.  <step id="STEP 2: CONSOLIDATE ANSWERS">
+    Consolidate all responses from the `ase-meta-search` *agents*
+    into a single response and output it without giving any further explanations.
+    </step>
+</flow>
+

package/plugin/skills/ase-meta-why/SKILL.md

@@ -0,0 +1,69 @@
+---
+name: ase-meta-why
+argument-hint: "<fact>"
+description: >
+    Five-Whys Root-Cause Analysis.
+user-invocable: true
+disable-model-invocation: false
+effort: medium
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Five-Whys Root-Cause Analysis
+=============================
+
+Your role is an *expert-level assistant*.
+
+<objective>
+Apply the *Five-Whys* *root-cause analysis* technique to investigate
+on the following problem:
+
+<problem>Why $ARGUMENTS?</problem>
+
+For this, iteratively ask "why" to drill down from symptoms to the root-cause.
+This helps to identify the fundamental reason behind a problem rather than just
+addressing surface-level symptoms.
+</objective>
+
+<flow>
+1.  <step id="STEP 1: PROBLEM">
+    State the problem statement.
+
+    <template>
+    &#x1F7E0; **PROBLEM**: <problem/>
+    </template>
+    </step>
+
+2.  <step id="STEP 2: ROOT-CAUSE ANALYSIS">
+    Find the root-cause of <problem/> by following the following iteration cycle.
+    Start with a <question/> equal to the <problem/>.
+
+    <for items="1 2 3 4 5">
+        Ask <question/> and document the <answer/> with the following template:
+        Don't stop at symptoms, keep digging for systemic issues.
+        Multiple root-causes may exist -- explore different branches.
+        Consider both technical, domain-specific, process-related or organizational causes.
+
+        <template>
+        &#x26AA; **WHY <item/>**: <answer/>
+        </template>
+
+        Then, for the next iteration let <question/> now be the last <answer/>.
+        The magic is NOT in exactly 5 "Why" -- you can <break/> the iteration
+        when you already reached the root-cause.
+    </for>
+    </step>
+
+3.  <step id="STEP 3: SOLUTION">
+    Validate the root-cause by working backwards the causality chain.
+    Propose a solution that addresses and solves the root-cause.
+    For the proposed solution, optionally directly propose corresponding source code changes.
+
+    <template>
+    &#x1F7E0; **SOLUTION**: <solution/>
+    </template>
+    </step>
+</flow>
+

package/plugin/skills/ase-task-delete/SKILL.md

@@ -0,0 +1,76 @@
+---
+name: ase-task-delete
+argument-hint: "[<id>]"
+description: >
+    Delete the current or given task plan.
+    Use when the user calls to "delete", "remove" or "clear" the
+    "task", "plan", "spec", or "specification".
+user-invocable: true
+disable-model-invocation: false
+effort: low
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-persona.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Delete a Task Plan
+==================
+
+<skill name="ase-task-delete">
+Delete a Task Plan
+</skill>
+
+Your role is an experienced, *expert-level assistant*.
+*Delete* the task plan.
+
+Procedure
+---------
+
+You *MUST* follow the following numbered items *strictly* *sequentially*!
+You *MUST* not skip any numbered item during processing!
+
+You *MUST* *NOT* output anything in this entire procedure, *except* when
+explicitly requested by this procedure via outputs based on a <template/>!
+
+1.  **Determine Task:**
+
+    1.  Set <id>$ARGUMENTS</id> initially.
+        Inherit the always existing <ase-task-id/> from the current context.
+        Do not output anything.
+
+    2.  <if condition="<id/> is empty">
+        Set <id><ase-task-id/></id>
+        Do not output anything.
+        </if>
+
+2.  **Perform Operation**:
+
+    1.  Call the `task_delete(id: <id/>)` tool of the `ase` MCP
+        service to delete the task plan content and set <text/> to the
+        `text` output field of this `task_delete` tool call. Do not
+        output anything related to this MCP tool call.
+
+        -   If <text/> starts with `ERROR:` or `WARNING:`:
+            Only output the following <template/>:
+
+            <template>
+            ⧉ **ASE**: ◉ task: **<id/>**, ▶ status: **<text/>**
+            </template>
+
+        -   If <text/> starts NOT with `ERROR:`:
+            Only output the following <template/>:
+
+            <template>
+            ⧉ **ASE**: ◉ task: **<id/>**, ▶ status: **task deleted**
+            </template>
+
+    2.  <if condition="<id/> is equal <ase-task-id/> AND <ase-task-id/> is not equal 'default'">
+        Set <ase-task-id>default</ase-task-id>. Call the `task_id(id:
+        <ase-task-id/>, session: <ase-session-id/>)` tool from the `ase`
+        MCP service to switch the task to the default task. Only output
+        the following <template/>:
+
+        <template>
+        ⧉ **ASE**: ◉ task: **<ase-task-id/>**, ▶ status: **task switched to default**
+        </template>
+        </if>