@rse/ase 0.9.50 → 0.9.52

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

@@ -20,9 +20,9 @@ allowed-tools:
 @${CLAUDE_SKILL_DIR}/../../meta/ase-dialog.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-proximity">
+<purpose name="ase-meta-proximity">
 Determine the Conceptual Proximity of a Topic
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-proximity"
@@ -37,37 +37,6 @@ relevant *child* topics:
 <topic><getopt-arguments/></topic>
 </objective>
 
-<define name="gather-facts">
-<if condition="<getopt-option-ground/> is equal `true`">
-Set <prompt>Search the Internet/Web and gather facts about
-<arg1/></prompt>.
-
-Then use the `ase-meta-search` skill in a sub-agent to gather facts with
-the following tool call and store the returned facts in the placeholder
-named `<arg2/>`:
-
-`Agent(
-    description: "Query Web Search Service",
-    subagent_type: "ase:ase-meta-search",
-    prompt: "<prompt/>",
-    run_in_background: false
-)`
-
-<if condition="the placeholder named `<arg2/>` contains no usable facts">
-Set the placeholder named `<arg2/>` to empty, so the determination below
-falls back to model knowledge, and output the following <template/>:
-
-<template>
-<ase-tpl-bullet-secondary/> **WARNING**: grounding found no usable facts -- falling back to model knowledge.
-</template>
-</if>
-</if>
-<else>
-Use the model's world knowledge and determine facts about <arg1/> and
-store those facts in the placeholder named `<arg2/>`.
-</else>
-</define>
-
 <flow>
 
 1.  <step id="STEP 1: Check Topic">
@@ -88,69 +57,59 @@ store those facts in the placeholder named `<arg2/>`.
     *REPEAT* the following sub-steps in a *LOOP* until either
     <getopt-option-loop/> is *not* equal `true` (then the loop runs
     exactly *once* and stops after rendering), or the user declines/cancels
-    the dialog in sub-step 5:
-
-    1.  *Determine Topic*:
-
-        Determine the canonical name of the central *topic* which is stored
-        in <topic/>.
-
-        <expand name="gather-facts"
-            arg1="the following topic: <topic/>"
-            arg2="facts-topic"></expand>
-
-        Ground the determination of the canonical name of the topic
-        <topic/> in the facts of <facts-topic/> and do not contradict
-        them. Update <topic/> accordingly.
-
-    2.  *Determine Proximity*:
-
-        Determine the *conceptual proximity* of the current <topic/>
-        along three *dimensions*:
-
-        -   **PARENT**:
-
-            The single most relevant *parent* topic (the broader topic
-            that <topic/> is a specialization of), which will be stored
-            in <parent/>.
+    the dialog in sub-step 4:
 
-            <expand name="gather-facts"
-                arg1="the PARENT topic (the broader topic that the given topic is a specialization of) of the following topic: <topic/>"
-                arg2="facts-parent"></expand>
+    1.  *Determine Proximity*:
 
-            Ground the determination of the canonical name of the parent
-            topic <parent/> in the facts of <facts-parent/> and do not
-            contradict them.
+        Set <parent/>, <sibling-1/> to <sibling-4/>, and <child-1/> to
+        <child-4/> to empty, so that a stale proximity of a previously
+        navigated topic can never leak into the rendering below.
 
-        -   **SIBLINGS**:
+        Set <prompt><topic/></prompt>.
 
-            The *four* most relevant *sibling* topics (topics on the
-            same level that share the same parent), which will be stored
-            in <sibling-1/> to <sibling-4/>.
-
-            <expand name="gather-facts"
-                arg1="the SIBLING topics (topics on the same level that share the same parent) of the following topic: <topic/>"
-                arg2="facts-siblings"></expand>
-
-            Ground the determination of the canonical names of the most
-            relevant sibling topics <sibling-1/> to <sibling-4/> in the
-            facts of <facts-siblings/> and do not contradict them.
-
-        -   **CHILDREN**:
-
-            The *four* most relevant *children* topics (narrower topics
-            that are specializations of <topic/>), stored in <child-1/>
-            to <child-4/>.
+        <if condition="<getopt-option-ground/> is equal `true`">
+        Set <prompt>GROUND <topic/></prompt>, so the agent grounds its
+        determination in Internet/Web facts instead of using model
+        knowledge only.
+        </if>
 
-            <expand name="gather-facts"
-                arg1="the CHILDREN topics (narrower topics that are specializations) of the following topic: <topic/>"
-                arg2="facts-children"></expand>
+        Determine the canonical name of the central *topic* and its
+        *conceptual proximity* along the three *dimensions* **PARENT**
+        (the single broader topic that <topic/> is a specialization of),
+        **SIBLINGS** (the four most relevant topics on the same level
+        that share the same parent), and **CHILDREN** (the four most
+        relevant narrower topics that are specializations of <topic/>) by
+        using the `ase-meta-proximity` agent in a sub-agent with the
+        following tool call:
+
+        `Agent(
+            description: "Determine Conceptual Proximity",
+            subagent_type: "ase:ase-meta-proximity",
+            prompt: "<prompt/>",
+            run_in_background: false
+        )`
+
+        Parse the returned labeled list and set <topic/> to the value of
+        its `TOPIC:` line, <parent/> to the value of its `PARENT:` line,
+        <sibling-1/> to <sibling-4/> to the values of its four `SIBLING:`
+        lines, and <child-1/> to <child-4/> to the values of its four
+        `CHILD:` lines.
+
+        <if condition="
+            the sub-agent returned no usable proximity OR
+            it did not yield a canonical topic, exactly one parent,
+            exactly four siblings, and exactly four children
+        ">
+        Determine the canonical topic name, the parent, the siblings, and
+        the children from model knowledge instead, and output the
+        following <template/>:
 
-            Ground the determination of the canonical names of the most
-            relevant children topics <child-1/> to <child-4/> in the
-            facts of <facts-children/> and do not contradict them.
+        <template>
+        <ase-tpl-bullet-secondary/> **WARNING**: proximity agent returned no usable result -- falling back to model knowledge.
+        </template>
+        </if>
 
-    3.  *Render Proximity*:
+    2.  *Render Proximity*:
 
         Output the result with the following <template/>, listing each
         proximity topic under its bullet-prefixed section header.
@@ -179,12 +138,12 @@ store those facts in the placeholder named `<arg2/>`.
         <ase-tpl-foot title="PROXIMITY TOPICS"/>
         </template>
 
-    4.  <if condition="<getopt-option-loop/> is not equal `true`">
+    3.  <if condition="<getopt-option-loop/> is not equal `true`">
         The loop runs only once in non-interactive mode: *break* out of
         the *loop* and stop processing without any further output.
         </if>
 
-    5.  *Navigate Proximity*:
+    4.  *Navigate Proximity*:
 
         In the following, you *MUST* *NOT* use your built-in
         <user-dialog-tool/> tool! Instead, you *MUST* just show a custom
@@ -207,7 +166,7 @@ store those facts in the placeholder named `<arg2/>`.
             CHILD-4:   ↓ <child-4/>
         </expand>
 
-        Check the tool <result/> and dispatch accordingly:
+        Check the <result/> and dispatch accordingly:
 
         -   If <result/> is `CANCEL`:
             *Break* out of the *loop* and stop processing without any

package/plugin/skills/ase-meta-proximity/help.md

@@ -20,11 +20,12 @@ topic that *topic* specializes), *SIBLINGS* (the four most relevant
 topics on the same level, sharing the same parent), and *CHILDREN* (the
 four most relevant narrower topics that specialize *topic*).
 
-By default the proximity is derived from model knowledge only. With the
-`--ground`/`-g` option, the skill first searches the Internet/Web for
-facts about the topic via the `ase-meta-search` skill (dispatched in a
-sub-agent, querying all available search backends) and grounds the
-determination in the found facts.
+The determination itself is delegated to the `ase-meta-proximity` agent,
+which is shared with the `ase-meta-quotes` skill. By default the
+proximity is derived from model knowledge only. With the `--ground`/`-g`
+option, the agent first searches the Internet/Web for facts about the
+topic, about its broader topic, and about its narrower topics, and
+grounds the determination in the found facts.
 
 Without the `--loop`/`-l` option, the skill determines and prints the
 proximity of the given topic once and then stops. With the `--loop`/`-l`
@@ -33,14 +34,14 @@ interactive dialog; selecting one of them makes it the new *current
 topic* and restarts the determination from the beginning, so the user
 can *navigate* the topic taxonomy up (parent), sideways (siblings), and
 down (children). Cancelling the dialog exits the loop. When `--loop` is
-combined with `--ground`, each newly selected topic is re-grounded via
-the `ase-meta-search` sub-agent before its proximity is re-determined.
+combined with `--ground`, each newly selected topic is re-grounded by
+the `ase-meta-proximity` sub-agent before its proximity is re-determined.
 
 ##  OPTIONS
 
 `--ground`|`-g`:
-    Ground the determination in Internet/Web facts gathered via the
-    `ase-meta-search` skill before determining the proximity. Without
+    Ground the determination in Internet/Web facts gathered by the
+    `ase-meta-proximity` agent before determining the proximity. Without
     this option, the determination is derived from model knowledge only.
     When combined with `--loop`, every newly navigated-to topic is
     re-grounded.
@@ -75,4 +76,4 @@ Internet/Web facts:
 
 ##  SEE ALSO
 
-[`ase-meta-eli5`](../ase-meta-eli5/help.md), [`ase-meta-search`](../ase-meta-search/help.md), [`ase-code-explain`](../ase-code-explain/help.md).
+[`ase-meta-quotes`](../ase-meta-quotes/help.md), [`ase-meta-eli5`](../ase-meta-eli5/help.md), [`ase-meta-search`](../ase-meta-search/help.md), [`ase-code-explain`](../ase-code-explain/help.md).

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

@@ -19,9 +19,9 @@ allowed-tools:
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-quorum">
+<purpose name="ase-meta-quorum">
 Query Multiple AIs for Quorum Answer
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-quorum"

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

@@ -0,0 +1,214 @@
+---
+name: ase-meta-quotes
+argument-hint: "[--help|-h] [--ground|-g] [--proximity|-p] [--count|-c <count>] <topic-keywords>"
+description: >
+    Find quotes for a set of topic keywords and place them into a
+    2x2 matrix, spanned by the presence of an author/origin and by
+    the literal containment of a keyword, optionally grounded in
+    Internet/Web facts and optionally widened to the conceptual
+    neighborhood of the topic. Use when the user wants "quotes",
+    "sayings", "aphorisms", or "citations" on a topic.
+user-invocable: true
+disable-model-invocation: false
+effort: high
+allowed-tools:
+    - "Agent"
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-control.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
+
+<purpose name="ase-meta-quotes">
+Find Quotes on a Topic
+</purpose>
+
+<expand name="getopt"
+    arg1="ase-meta-quotes"
+    arg2="--ground|-g --proximity|-p --count|-c=8">
+    $ARGUMENTS
+</expand>
+
+<objective>
+*Find* quotes for the following topic keywords:
+<keywords><getopt-arguments/></keywords>
+</objective>
+
+<flow>
+
+1.  <step id="STEP 1: Sanity Check Usage">
+
+    1.  <if condition="<keywords/> is empty">
+        Only output the following <template/> and then immediately *STOP*
+        processing the entire current skill:
+
+        <template>
+        ⧉ **ASE**: ✪ skill: **ase-meta-quotes**, ▶ ERROR: expected a `<topic-keywords>` argument
+        </template>
+        </if>
+
+    2.  Set <quotes></quotes> (set to empty).
+
+    3.  Determine the maximum total number of *quotes* to surface: set
+        <count/> to <getopt-option-count/>; if <getopt-option-count/> is
+        *non-numeric* or *less than or equal to 0*, use the default *8*
+        instead.
+
+    </step>
+
+2.  <step id="STEP 2: Harvest Quotes">
+
+    1.  Determine *quotes* -- sayings, aphorisms, maxims, proverbs, and
+        citations -- which are about the topic <keywords/>, and store them in
+        <quotes/>. Per quote, record its *text*, its *author* (a named person
+        or organization, if any is known), its *origin* (a named work,
+        standard, or document, if any is known), and <keywords/> as its
+        *source topic*.
+
+    2.  <if condition="<getopt-option-ground/> is equal `true`">
+
+        *Additionally* -- and never *instead* -- gather quotes from the
+        Internet/Web by using the `ase-meta-search` skill in a sub-agent
+        with the following tool call:
+
+        `Agent(
+            description: "Query Web Search Service",
+            subagent_type: "ase:ase-meta-search",
+            prompt: "Search the Internet/Web and gather quotes about the following topic: <keywords/>",
+            run_in_background: false
+        )`
+
+        Merge the returned quotes into <quotes/>, deduplicating quotes which
+        differ only in punctuation, capitalization, or attribution wording,
+        and remember for every quote whether the search *confirmed* its exact
+        wording and attribution.
+
+        <if condition="the sub-agent returned no usable quotes">
+        Output the following <template/> and continue with the model
+        knowledge only:
+
+        <template>
+        <ase-tpl-bullet-secondary/> **WARNING**: grounding found no usable quotes -- falling back to model knowledge.
+        </template>
+        </if>
+
+        </if>
+
+    </step>
+
+3.  <step id="STEP 3: Widen Topic via Proximity" condition="<getopt-option-proximity/> is equal `true`">
+
+    1.  Set <prompt><keywords/></prompt>.
+
+    2.  <if condition="<getopt-option-ground/> is equal `true`">
+        Set <prompt>GROUND <keywords/></prompt>, so the agent grounds
+        its determination in Internet/Web facts instead of using model
+        knowledge only.
+        </if>
+
+    3.  Determine the *conceptual neighborhood* of <keywords/> by using the
+        `ase-meta-proximity` agent in a sub-agent with the following
+        tool call:
+
+        `Agent(
+            description: "Determine Conceptual Proximity",
+            subagent_type: "ase:ase-meta-proximity",
+            prompt: "<prompt/>",
+            run_in_background: false
+        )`
+
+    4.  <if condition="the sub-agent returned no usable neighborhood">
+        Output the following <template/>, *SKIP* the remaining sub-steps
+        of this step, and continue with the quotes harvested in STEP 2
+        only:
+
+        <template>
+        <ase-tpl-bullet-secondary/> **WARNING**: proximity agent returned no usable result -- keeping the narrow topic only.
+        </template>
+        </if>
+
+    5.  Parse the returned labeled list and set <neighborhood/> to the values
+        of its `PARENT:` line (the *broader* topic), of its four
+        `SIBLING:` lines (the *same-level* topics), and of its four
+        `CHILD:` lines (the *narrower* topics).
+
+    6.  Harvest quotes for *each* topic of <neighborhood/> exactly as in
+        STEP 2 (Harvest Quotes), record the contributing neighborhood
+        topic as the *source topic* of each of those quotes, and merge
+        the results into <quotes/>.
+
+    </step>
+
+4.  <step id="STEP 4: Classify and Render Quotes">
+
+    1.  *Classify Quotes*:
+
+        Classify every quote of <quotes/> along two *orthogonal* axes:
+
+        -   **ATTRIBUTION**:
+            A quote is `ATTRIBUTED` if a named *author* and/or a named
+            *origin* is known for it, and `ANONYMOUS` otherwise.
+
+        -   **LITERALNESS**:
+            A quote is `LITERAL` if its text contains at least one of
+            the topic keywords of <keywords/> as a *whole word* --
+            matched *case-insensitively* and tolerating *inflections*
+            (e.g. `architect` and `architectural` match the keyword
+            `architecture`), but *never* as a mere *substring* (e.g.
+            `art` does *not* match `architecture`). A quote is
+            `THEMATIC` otherwise.
+
+        Both axes span the four *quadrants*:
+
+        -   `Q1` (`ATTRIBUTED` and `LITERAL`)
+        -   `Q2` (`ATTRIBUTED` and `THEMATIC`)
+        -   `Q3` (`ANONYMOUS`  and `LITERAL`)
+        -   `Q4` (`ANONYMOUS`  and `THEMATIC`)
+
+    2.  Finally, reduce <quotes/> to at most <count/> quotes in *total*,
+        distributed as evenly as possible across the four quadrants
+        and preferring the most relevant and most well-known quote per
+        quadrant.
+
+    3.  *Render Quotes*:
+
+        Render every quote on its own `○`-prefixed line, with the
+        following *suffixes* appended in this order:
+
+        -   ` — *<author/>*, <origin/>` in the two `ATTRIBUTED` quadrants,
+            omitting whichever of <author/> and <origin/> is unknown.
+
+        -   ` [from proximity: `<source-topic/>`]` if the quote was
+            harvested for a *neighborhood* topic in STEP 3 rather than for
+            <keywords/> itself.
+
+        -   ` *(unverified)*` if the exact wording or the attribution of
+            the quote could not be established with confidence -- but
+            omit this marker for a quote whose wording and attribution
+            the Internet/Web search of STEP 2.2 confirmed -- including
+            its re-application for a neighborhood topic in STEP 3.
+
+        Render the single line `○   (none)` for a quadrant without any
+        quote. Output the result with the following <template/>:
+
+        <template>
+        <ase-tpl-head title="QUOTES"/>
+
+        ●   **Q1 - ATTRIBUTED / LITERAL**:
+        ○   [...]
+
+        ●   **Q2 - ATTRIBUTED / THEMATIC**:
+        ○   [...]
+
+        ●   **Q3 - ANONYMOUS / LITERAL**:
+        ○   [...]
+
+        ●   **Q4 - ANONYMOUS / THEMATIC**:
+        ○   [...]
+
+        <ase-tpl-foot title="QUOTES"/>
+        </template>
+
+    </step>
+
+</flow>

package/plugin/skills/ase-meta-quotes/help.md

@@ -0,0 +1,109 @@
+
+##  NAME
+
+`ase-meta-quotes` - Find Quotes on a Topic
+
+##  SYNOPSIS
+
+`ase-meta-quotes`
+    [`--help`|`-h`]
+    [`--ground`|`-g`]
+    [`--proximity`|`-p`]
+    [`--count`|`-c` *count*]
+    *topic-keywords*
+
+##  DESCRIPTION
+
+The `ase-meta-quotes` skill finds *quotes* -- sayings, aphorisms,
+maxims, proverbs, and citations -- for the supplied *topic-keywords* and
+places them into a *2x2 matrix*, reported as four labeled *quadrant*
+sections.
+
+Per quote, the skill records its *text*, its *author* (a named person or
+organization, if known), its *origin* (a named work, standard, or
+document, if known), and the topic it was harvested for (its *source
+topic*).
+
+The first dimension of the matrix is the *attribution* of a quote: a
+quote is `ATTRIBUTED` when a named author and/or a named origin is known
+for it, and `ANONYMOUS` otherwise. The second dimension is the
+*literalness* of a quote: a quote is `LITERAL` when its text contains at
+least one of the *topic-keywords* as a whole word -- matched
+case-insensitively and tolerating inflections (e.g. `architect` and
+`architectural` match the keyword `architecture`), but never as a mere
+substring (e.g. `art` does not match `architecture`) -- and `THEMATIC`
+otherwise. The two dimensions span the quadrants `Q1`
+(`ATTRIBUTED`/`LITERAL`), `Q2` (`ATTRIBUTED`/`THEMATIC`), `Q3`
+(`ANONYMOUS`/`LITERAL`), and `Q4` (`ANONYMOUS`/`THEMATIC`). A quadrant
+without any quote is rendered as `(none)`.
+
+Every quote is rendered on its own line and carries, where applicable,
+the suffixes `— <author>, <origin>` (in the two `ATTRIBUTED`
+quadrants, omitting whichever part is unknown), `[from proximity:
+<source-topic>]` (when the quote was contributed by a neighborhood topic
+under `--proximity` instead of by *topic-keywords* itself), and
+`(unverified)` (when the exact wording or the attribution could not be
+established with confidence). The `(unverified)` marker is dropped as
+soon as the Internet/Web search under `--ground` confirms the wording and
+the attribution of the quote.
+
+##  OPTIONS
+
+`--ground`|`-g`:
+    Gather quotes from the Internet/Web via the `ase-meta-search` skill
+    (dispatched in a sub-agent, querying all available search backends)
+    *in addition to* -- and never *instead* of -- the model knowledge,
+    merge them into the harvest while deduplicating quotes which differ
+    only in punctuation, capitalization, or attribution wording, and use
+    the search results to confirm the wording and the attribution of the
+    found quotes. When combined with `--proximity`, the determination of
+    the conceptual neighborhood is grounded in Internet/Web facts as
+    well. Should the search return no usable quotes, a warning is emitted
+    and the skill falls back to the model knowledge only. Without this
+    option, the quotes are derived from model knowledge only.
+
+`--proximity`|`-p`:
+    Widen the harvest beyond the given topic to its *conceptual
+    neighborhood* -- the parent (more general) topic, the four sibling
+    (same-level) topics, and the four child (more specialized) topics --
+    as determined by the `ase-meta-proximity` agent, which is shared with
+    the `ase-meta-proximity` skill. Each of these nine neighborhood
+    topics is harvested exactly like the given topic itself. Quotes
+    contributed by a neighborhood topic carry that topic in brackets and
+    mostly land in the `THEMATIC` quadrants, which is exactly where the
+    widening pays off. Should the agent return no usable neighborhood, a
+    warning is emitted and the skill continues with the narrow topic
+    only. Without this option, only the given topic is harvested.
+
+`--count`|`-c` *count*:
+    Maximum total number of quotes across all four quadrants (default:
+    `8`, i.e. about two quotes per quadrant). The quotes are distributed
+    as evenly as possible across the quadrants, preferring the most
+    relevant and most well-known quote per quadrant. A non-numeric value
+    or a value less than or equal to `0` falls back to the default.
+
+##  ARGUMENTS
+
+*topic-keywords*:
+    The topic keywords the quotes are searched for. They act both as the
+    search query and as the keyword set of the *literalness* dimension of
+    the matrix.
+
+##  EXAMPLES
+
+Find quotes from model knowledge:
+
+```text
+❯ /ase-meta-quotes software architecture
+```
+
+Find a wider set of quotes, grounded in Internet/Web facts and widened
+to the conceptual neighborhood of the topic:
+
+```text
+❯ /ase-meta-quotes --ground --proximity --count 12 technical debt
+```
+
+##  SEE ALSO
+
+[`ase-meta-proximity`](../ase-meta-proximity/help.md), [`ase-meta-search`](../ase-meta-search/help.md), [`ase-meta-eli5`](../ase-meta-eli5/help.md).

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

@@ -19,9 +19,9 @@ allowed-tools:
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-review">
+<purpose name="ase-meta-review">
 Review Staged Changes
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-review"

package/plugin/skills/ase-meta-review/help.md

@@ -59,9 +59,10 @@ before it is committed.
     Set the *severity floor* (default `LOW`): findings below the chosen
     threshold are silently suppressed, ordered `LOW` < `MEDIUM` <
     `HIGH`. The default `LOW` keeps all findings; `ACCEPTED` findings are
-    never suppressed. The floor only affects the rendered findings table,
-    not the overall *verdict*, which is always derived from all findings
-    before the floor is applied.
+    never suppressed. Surviving findings are rendered in *descending
+    severity* order `HIGH`, `MEDIUM`, `LOW`, `ACCEPTED`. The floor only
+    affects the rendered findings table, not the overall *verdict*, which
+    is always derived from all findings before the floor is applied.
 
 ##  ARGUMENTS
 

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

@@ -19,9 +19,9 @@ allowed-tools:
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-search">
+<purpose name="ase-meta-search">
 Search the Internet/Web
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-search"

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

@@ -14,9 +14,9 @@ effort: xhigh
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-steelman">
+<purpose name="ase-meta-steelman">
 Build the "Steelman" Argument
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-steelman"

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

@@ -12,9 +12,9 @@ effort: high
 @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
 @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
 
-<skill name="ase-meta-why">
+<purpose name="ase-meta-why">
 Five-Whys Root-Cause Analysis
-</skill>
+</purpose>
 
 <expand name="getopt"
     arg1="ase-meta-why"