@rse/ase 0.0.48 → 0.0.50

package/plugin/skills/ase-docs-proofread/SKILL.md

@@ -48,96 +48,25 @@ documents.
     ⚪ **PROOFREADING INVESTIGATION**
     </template>
 
-    Then dispatch the investigation to a *sub-agent* via the `Agent`
+    Dispatch the investigation to a *sub-agent* via the `Agent`
     tool so that *no* investigation details leak into the user-visible
     transcript. The sub-agent performs the silent reading and checking;
     only its final structured return value is consumed here.
 
-    Invoke the `Agent` tool *exactly once* with:
-
-    -   `subagent_type`: `general-purpose`
-    -   `description`: `Proofread Investigation`
-    -   `prompt`: a *self-contained* briefing, instructing the sub-agent to:
-
-        1.  Use the `Read` tool to read all document files referenced
-            by <getopt-arguments/>.
-
-        2.  Check the contained texts *only* for the following problem
-            types:
-
-            - **Spelling**
-            - **Punctuation**
-            - **Grammar**
-
-            Do *NOT* flag stylistic preferences, Markdown formatting
-            choices, code/identifiers, XML/template tags, technical
-            terms, intentional capitalization, list/heading style, or
-            anything inside fenced code blocks or backtick spans. Be
-            conservative — only report clear, objective errors.
-
-            For each found problem:
-
-            -   Set <type/> to the string `SPELLING`, `PUNCTUATION`, or
-                `GRAMMAR`, indicating the problem type.
-
-            -   Set <file/> to the *relative* filename path of the document.
-                Set <line/> to the numeric 1-based line number in the
-                document.
-
-            -   Set <old-text/> to the lines of the old text which
-                should be changed. Set <new-text/> to the lines of the
-                new text which will be changed.
-
-            -   Set <description/> to an ultra-brief and concise
-                Markdown-formatted description of the problem with
-                a hint of what is wrong and why it is wrong. In
-                this description, mark up all referenced verbatim
-                words <words/> from <old-text/> or <new-text/> as
-                quoted strings containing monospaced text with
-                Markdown based on the following <template/>:
-                <template>"`<words/>`"</template>.
-
-            -   Set <context-before/> to exactly *up to two* lines of
-                *unchanged* text context which occurs in the document
-                directly *before* <old-text/>, i.e., the lines (<line/> -
-                2) and (<line/> - 1). Reduce to just one line (<line/> -
-                1) if <old-text/> is the second line of the document. Set
-                <context-before/> to empty if <old-text/> is the first line in
-                the document.
-
-            -   Set <context-after/> to exactly *up to two* lines of
-                *unchanged* text content which occurs in the document
-                directly *after* <old-text/> the lines (<line/> + <n/> + 1)
-                and (<line/> + <n/> + 2), where <n/> is the number of lines
-                in <old-text/>. Reduce to just one line (<line/> + <n/> + 1)
-                if <old-text/> is the second-last line of the document. Set
-                <context-after/> to empty if <old-text/> is the last line in
-                the document.
-
-        3.  Return *exclusively* a single fenced JSON block (no prose,
-            no preamble, no summary) of the following shape:
-
-            ```json
-            [
-                {
-                    "type":           <type/>,
-                    "file":           <file/>,
-                    "line":           <line/>,
-                    "description":    <description/>,
-                    "context_before": <context-before/>,
-                    "old_text":       <old-text/>,
-                    "new_text":       <new-text/>,
-                    "context_after":  <context-after/>
-                },
-                [...]
-            ]
-            ```
+    For this, invoke *exactly once* the tool:
 
-        4.  You *MUST* *NOT* propose, apply, or render any document
-            changes itself.
+    ```text
+        Agent(
+            name:          "ase:ase-docs-proofread",
+            description:   "Proofread Investigation",
+            subagent_type: "ase:ase-docs-proofread",
+            mode:          "plan",
+            prompt:        <getopt-arguments/>
+        )
+    ```
 
-    Parse the JSON array from the sub-agent's return value and set
-    <problems/> to that list.
+    Parse the single result message of the `Agent` tool as a JSON array
+    and set <problems/> to that list.
 
     You *MUST* *NOT* output anything at all in this step 1 beyond the
     single `Agent` tool invocation.
@@ -169,11 +98,9 @@ documents.
 
 3.  <step id="STEP 3: Correction">
 
-    1.  You *MUST* activate the auto-approve gate for the `Edit` tool
-        by setting the session-scoped `agent.skill` configuration value
-        to this skill's name via the MCP tool call `config_set(key:
-        "agent.skill", val: "ase-docs-proofread", scope:
-        "session:<ase-session-id/>")` of the `ase` service.
+    1.  You *MUST* call the MCP tool `config_set(key: "agent.skill", val:
+        "ase-docs-proofread", scope: "session:<ase-session-id/>")` of the
+        `ase` MCP server.
 
     2.  Iterate over all problems:
 
@@ -230,7 +157,6 @@ documents.
             <expand name="user-dialog">
                 CORRECTION: How would you like to proceed with this proposed correction?
                 ACCEPT: Apply the proposed correction.
-                REFINE: Discard this proposed correction and generate a new one.
                 REJECT: Skip this proposed correction.
             </expand>
 
@@ -253,15 +179,12 @@ documents.
 
                 </if>
 
-            -   <if condition="
-                    <result/> starts with 'REFINE' or
-                    <result/> starts with 'OTHER'
-                ">
+            -   <if condition="<result/> starts with 'OTHER'">
 
-                Generate a *new* proposal for the *same* <item/>
-                (incorporating the user's free-text hint from <result/> if
-                provided via the "OTHER" prefix) and loop back to substep 2
-                of this iteration. There is *no* cap on refinement rounds —
+                Generate a *new* proposal for the *same* <item/>,
+                incorporating the user's free-text hint from <result/>
+                after the "OTHER:" prefix, and loop back to substep 2 of
+                this iteration. There is *no* cap on refinement rounds —
                 keep refining until the user picks `ACCEPT` or `REJECT`.
 
                 </if>
@@ -279,13 +202,16 @@ documents.
 
         </for>
 
-    3.  After the iteration has finished, you *MUST* clear the auto-approve
-        gate via the call MCP tool call `config_delete(key:
-        "agent.skill", scope: "session:<ase-session-id/>")` of the `ase`
-        service.
+    3.  You *MUST* call the MCP tool `config_delete(key: "agent.skill",
+        scope: "session:<ase-session-id/>")` of the `ase` MCP server.
 
     4.  You *MUST* *NOT* output any further additional explanations or
-        summaries at the end of this skill processing.
+        summaries at the end of this skill processing, except for the
+        following final <template/>:
+
+        <template>
+        &#x26AA; **PROOFREAD FINISHED**
+        </template>
 
     </step>
 

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

@@ -2,64 +2,48 @@
 name: ase-meta-chat
 argument-hint: "<llm> <query>"
 description: >
-    Query foreign LLM for Chat.
+    Query foreign LLM for chat.
     Use this skill if a foreign LLM like OpenAI ChatGPT, Google Gemini,
     DeepSeek or xAI Grok should be queried with a single chat message.
 user-invocable: true
 disable-model-invocation: false
-context: fork
 effort: low
 allowed-tools:
-    - "mcp__chat-openai-chatgpt"
-    - "mcp__chat-google-gemini"
-    - "mcp__chat-deepseek"
-    - "mcp__chat-xai-grok"
-    - "Task"
     - "Agent"
 ---
 
 @${CLAUDE_SKILL_DIR}/../../meta/ase-control.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 
-Query Foreign LLMs for Chat
-===========================
+Query Foreign LLM for Chat
+==========================
 
 <skill name="ase-meta-chat">
-Query Foreign LLMs for Chat
+Query Foreign LLM for Chat
 </skill>
 
+<role>
 Your role is to act as a proxy to query a foreign LLM for a single chat message.
+</role>
 
 <objective>
 Query foreign LLM for: <query>$ARGUMENTS</query>
 </objective>
 
-<flow>
-
-1.  <step id="STEP 1: Select LLMs">
-    Use the *first word* of the following <query/> for selecting the
-    foreign LLMs to query, and their corresponding MCP servers, from the
-    following list:
-
-    - **OpenAI ChatGPT**: via MCP server `chat-openai-chatgpt`
-    - **Google Gemini**:  via MCP server `chat-google-gemini`
-    - **DeepSeek**:       via MCP server `chat-deepseek`
-    - **xAI Grok**:       via MCP server `chat-xai-grok`
-    </step>
-
-2.  <step id="STEP 2: Spawn Agents">
-    Spawn a *sub-task* with the companion `ase-meta-chat` *agent* (and
-    not this *skill*, but the agent of the same name) for the selected
-    foreign LLMs, and pass the *second and all remaining* words of the
-    following <query/> as the query for the selected LLM.
-    </step>
-
-3.  <step id="STEP 3: Return Responses">
-    Return the *plain response* of the `ase-meta-chat` agent 1:1 and
-    *without any modifications*. Especially, do *NOT* add or remove any
-    text from the agent response on your own and do not interpret the
-    result in any way.
-    </step>
-
-</flow>
+1.  You *MUST* *NOT* output anything in this step.
+    Just call the underlying agent with the following tool:
+
+    ```text
+        Agent(
+            name:          "ase:ase-meta-chat",
+            description:   "Query Foreign LLM for Chat",
+            subagent_type: "ase:ase-meta-chat",
+            prompt:        <query/>
+        )
+    ```
+
+2.  Output the *plain response* of the `ase:ase-meta-chat` agent
+    *verbatim* and *without any modifications*. Especially, do *NOT* add or
+    remove any text from the agent response on your own and do not interpret
+    the result in any way.
 

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

@@ -248,7 +248,7 @@ multi-*criteria* decision matrix.
         ⚠ *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
+    -   Otherwise (<percentage/> is greater than or equal to 0.10), output
         just the following <template/> and do not output anything else:
 
         <template>

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

@@ -6,7 +6,7 @@ description: >
     The <persona> can be either the decorative, eloquent, and explaining "writer",
     the concise, factual, and accurate "engineer" (default),
     the brief, factual, and abbreviating "telegrapher",
-    the terse, rough and stuttering "caveman".
+    the terse, rough, and stuttering "caveman".
     Use when user says "persona <persona>" or "be <persona>".
 user-invocable: true
 disable-model-invocation: false

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

@@ -21,7 +21,9 @@ Query Multiple AIs for Quorum Answer
 Query Multiple AIs for Quorum Answer
 </skill>
 
+<role>
 Your role is an *expert-level assistant*.
+</role>
 
 <objective>
 Find a *quorum answer* on an arbitrary question,
@@ -29,11 +31,12 @@ 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>
+1.  <step id="STEP 1: Preview Own Answer">
+
+    Prepare the LLM query by setting <query/> to the following <template/>:
+
+    <template>
     $ARGUMENTS.
     Please respond with facts and very concise and brief only,
     usually with just 1 to 7 corresponding bullet points and with short sentences.
@@ -41,55 +44,80 @@ by querying *multiple* AIs for an *optimal consensus*.
     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>
+    </template>
 
-2.  <step id="STEP 2: Preview Own Answer">
-    Show your own answer as a sneak preview.
-    For this, use the following output <template/>:
+    For yourself (Anthropic Claude), first answer this <query/> in
+    advance yourself by showing your own answer to the query as a sneak
+    preview. For this, output the following <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:
+2.  <step id="STEP 2: Query Foreign AIs">
+
+    <define name="agent">
+    Call the `Agent` tool:
+
+    ```text
+        Agent(
+            name:          "ase:ase-meta-chat",
+            description:   "Query Foreign LLM: <arg1/>",
+            subagent_type: "ase:ase-meta-chat",
+            prompt:        "<arg2/> <query/>"
+        )
+    ```
+
+    </define>
+
+    <expand name="agent" arg1="OpenAI ChatGPT" arg2="chatgpt"></expand>
+    <expand name="agent" arg1="Google Gemini"  arg2="gemini"></expand>
+    <expand name="agent" arg1="DeepSeek"       arg2="deepseek"></expand>
+    <expand name="agent" arg1="xAI Grok"       arg2="grok"></expand>
 
-    - OpenAI ChatGPT: `chat-openai-chatgpt`
-    - Google Gemini:  `chat-google-gemini`
-    - DeepSeek:       `chat-deepseek`
-    - xAI Grok:       `chat-xai-grok`
+    You *MUST* *NOT* output anything in this step.
 
-    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.
+3.  <step id="STEP 3: Summarize Responses">
+
+    Agents which returned a response with an `ERROR:` prefix are
+    silently skipped and are treatd as not available.
+
+    Summarize all responses, of both yourself and all available agents
+    with just 1 to 7 corresponding bullet points and with short
+    sentences.
+
+    You *MUST* *NOT* output anything in this step.
+
     </step>
 
-5.  <step id="STEP 5: Determine Consensus Rating">
+4.  <step id="STEP 4: 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 the *total number of
     responders* (yourself plus all available foreign 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,
+    If not all AIs agree, determine <disagreement/> information,
     formatted as `(disagreement: <ai/>, <ai/>, [...])` where <ai/> is a
     name of an AI which disagreed with the consensus. Else, if all AIs
     agree, set <disagreement></disagreement>.
+
+    You *MUST* *NOT* output anything in this step.
+
     </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/>:
+5.  <step id="STEP 5: 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**:
@@ -123,6 +151,9 @@ by querying *multiple* AIs for an *optimal consensus*.
     </template>
 
     In this output, remove the sections of those AIs which were not available.
+    You *MUST* *NOT* output any further explanations yourself.
+
     </step>
+
 </flow>
 

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

@@ -10,9 +10,7 @@ effort: low
 allowed-tools:
     - "mcp__perplexity__perplexity_search"
     - "mcp__brave__brave_web_search"
-    - "WebSearch"
-    - "WebFetch"
-    - "Task"
+    - "Agent"
 ---
 
 @${CLAUDE_SKILL_DIR}/../../meta/ase-control.md
@@ -25,7 +23,9 @@ Search the Internet/Web
 Search the Internet/Web
 </skill>
 
+<role>
 Your role is an expert-level *web specialist*.
+</role>
 
 <objective>
 Your objective is to *search* the *Internet*/*Web* for the following query:
@@ -33,20 +33,46 @@ Your objective is to *search* the *Internet*/*Web* for the following 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*.
+1.  <step id="STEP 1: Query Search Services">
+
+    <define name="agent">
+    ```text
+        Agent(
+            name:          "ase:ase-meta-search",
+            description:   "Query Web Search Service",
+            subagent_type: "ase:ase-meta-search",
+            prompt:        <content/>
+        )
+    ```
+    </define>
+
+    If the MCP tool `mcp__perplexity__perplexity_search` is available, call:
+    <expand name="agent">
+        Call the MCP tool `mcp__perplexity__perplexity_search(query: "<query/>")`
+    </expand>
+
+    If the MCP tool `mcp__brave__brave_web_search` is available, call:
+    <expand name="agent">
+        Call the MCP tool `mcp__brave__brave_web_search(query: "<query/>")`
+    </expand>
+
+    <expand name="agent">
+        Call the tool `WebSearch(query: "<query/>")`
+    </expand>
 
-    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.
+2.  <step id="STEP 2: Consolidate Search Answers">
+
+    Consolidate all responses from the `ase:ase-meta-search` agents
+    calls above into a single response and output it without giving any
+    further explanations.
+
+    For the consolidation, do *NOT* remove any orginal information,
+    just *MERGE* all overlapping information.
+
     </step>
+
 </flow>
 

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

@@ -0,0 +1,92 @@
+---
+name: ase-task-rename
+argument-hint: "[<old>] <new>"
+description: >
+    Rename the current or given task plan.
+    Use when the user calls to "rename", "move" or "relabel" the
+    "task", "plan", "spec", or "specification".
+user-invocable: true
+disable-model-invocation: false
+effort: low
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-control.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+
+Rename a Task Plan
+==================
+
+<skill name="ase-task-rename">
+Rename a Task Plan
+</skill>
+
+Your role is an experienced, *expert-level assistant*.
+*Rename* 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.  Parse <arguments>$ARGUMENTS</arguments> into a whitespace-separated
+        list of tokens. Inherit the always existing <ase-task-id/> from
+        the current context. Do not output anything.
+
+    2.  <if condition="<arguments/> contains two tokens">
+        Set <old/> to the first token of <arguments/>.
+        Set <new/> to the second token of <arguments/>.
+        Do not output anything.
+        </if>
+
+    3.  <if condition="<arguments/> contains exactly one token">
+        Set <old><ase-task-id/></old>.
+        Set <new/> to the single token of <arguments/>.
+        Do not output anything.
+        </if>
+
+    4.  <if condition="<arguments/> is empty OR contains more than two tokens">
+        Only output the following <template/> and then immediately
+        *STOP* processing the entire current skill:
+
+        <template>
+        ⧉ **ASE**: ☻ skill: **ase-task-rename**, ▶ ERROR: expected `[<old>] <new>` arguments
+        </template>
+        </if>
+
+2.  **Perform Operation**:
+
+    1.  Call the `task_rename(old: <old/>, new: <new/>)` tool of the
+        `ase` MCP service to rename the task plan and set <text/> to the
+        `text` output field of this `task_rename` 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: **<old/>**, ▶ status: **<text/>**
+            </template>
+
+        -   If <text/> starts NOT with `ERROR:` and NOT with `WARNING:`:
+            Only output the following <template/>:
+
+            <template>
+            ⧉ **ASE**: ◉ task: **<new/>**, ▶ status: **task renamed**
+            </template>
+
+    2.  <if condition="<old/> is equal <ase-task-id/>">
+        Set <ase-task-id><new/></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 renamed task. Only output
+        the following <template/>:
+
+        <template>
+        ⧉ **ASE**: ◉ task: **<ase-task-id/>**, ▶ status: **task switched**
+        </template>
+        </if>