@rse/ase 0.9.47 → 0.9.49

package/plugin/meta/ase-getopt.md

@@ -76,6 +76,15 @@ set placeholders into the context as a side-effect.
     ⧉ **ASE**: ✪ skill: **<getopt-skill/>**, ▶ ERROR: option parsing failed: **<text/>**
     </template>
 
+    Directly *after* this error <template/>, and *before* stopping, give
+    the corrective hint by expanding the following (which, depending on
+    the configured <ase-guidance-level/>, may expand into nothing and
+    hence emit no output at all):
+
+    <ase-tpl-hint level="verbose">
+    Run `/<getopt-skill/> --help` for the accepted options and arguments of this skill.
+    </ase-tpl-hint>
+
 5.  **Parsing JSON Result**:
     The tool returned a single `text` content payload containing JSON.
     Parse this JSON in <text/> now into <getopt-result/> by recognizing
@@ -100,7 +109,8 @@ set placeholders into the context as a side-effect.
     whose long name starts with `int-`.
 
 7.  **Display Results**:
-    Just output the following <template/>:
+    If <ase-guidance-level/> is equal to `normal` or `verbose`,
+    output the following <template/>:
 
     <template>
     ⧉ **ASE**: ✪ skill: **<getopt-skill/>**, ▶ options: <getopt-info/>

package/plugin/meta/ase-skill.md

@@ -214,15 +214,17 @@ Skill Identification
 
     In case <skill/> later becomes *not* empty by defining it as <skill
     name="<name/>"><body/></skill>, set <skill-name><name/></skill-name>
-    (set skill name to name), and then (but only if `$1` is *NOT* equal
-    to `-h` or `--help`) you *MUST* once output the following output
+    (set skill name to name), and then (but only if ((`$1` is *NOT* equal
+    to `-h` or `--help`) *AND* (<ase-guidance-level/> is equal to `normal`
+    or `verbose`))) you *MUST* once output the following output
     <template/>:
 
     <template>
     ⧉ **ASE**: ✪ skill: **<skill-name/>**, ✦ purpose: **<skill/>**, ▶ status: **skill started**
     </template>
 
-    Later (but only if `$1` is *NOT* equal to `-h` or `--help`), once
+    Later (but only if ((`$1` is *NOT* equal to `-h` or `--help`) *AND*
+    (<ase-guidance-level/> is equal to `normal` or `verbose`))), once
     this skill finally will stop processing, you *MUST* once output the
     following output <template/>:
 
@@ -231,7 +233,8 @@ Skill Identification
     </template>
 
 -   *IMPORTANT*: Set <objective></objective> (set to empty).
-    Then, in case <objective/> later becomes *not* empty,
+    Then, in case <objective/> later becomes *not* empty
+    (but only if <ase-guidance-level/> is equal to `normal` or `verbose`),
     you *MUST* once output the following output <template/>:
 
     <template>
@@ -292,6 +295,65 @@ Artifact Boxing Transparency
     deterministically skip or suppress and *win* over the implicit decisions
     above. Where no such branches exist, the decisions above apply.
 
+Guidance Hint Level
+-------------------
+
+-   *IMPORTANT*: The *guidance* of the agent (indicated by
+    <ase-guidance-level/> configuration) classifies how many
+    *unsolicited hints* -- pointers to the available *ASE* skills, to
+    their options, and to the recommended next operations, which the
+    user did *not* explicitly ask for -- you *MUST* emit in addition to
+    the requested result.
+
+-   *IMPORTANT*: Guidance modulates only the *hints*, per the four
+    levels below. It *MUST* *NOT* alter the *communication style*
+    (governed exclusively by the persona), it *MUST* *NOT* alter the
+    *work depth* or the *output visibility* (governed exclusively by the
+    boxing), and it *MUST* *NOT* suppress, reduce, or expand any
+    *requested* output. In particular, guidance *MUST* *NOT* apply to:
+
+    -   the *manual page* emitted for `-h`/`--help`, which is
+        explicitly requested,
+    -   the *skill catalog*, *manual page*, and *command proposal*
+        results of the `ase-help-skill` and `ase-help-intent` skills,
+        which are explicitly requested,
+    -   the *option affordance* lines of a user dialog (e.g. `Please
+        choose *one* option by typing ...`), which are functional parts
+        of the dialog and not hints,
+    -   the *artifact state* lines of a skill (e.g. `plan loaded`, `plan
+        created`, `task given`) and every `ERROR:` line, which carry
+        state rather than pointers.
+
+    The *skill identification* chrome (the `skill started`, `skill
+    finished`, and `objective` lines) is the sole exception: it is
+    *decoration* rather than result, so it is gated on `normal` and
+    `verbose` directly in the `Skill Identification` section above.
+
+-   *IMPORTANT*: Every hint carries a *level* -- `minimal`, `normal`,
+    or `verbose` -- and is emitted only if the configured
+    <ase-guidance-level/> *includes* that level:
+
+    -   If <ase-guidance-level/> is `none`:
+        -   Hints emitted: *none*
+    -   If <ase-guidance-level/> is `minimal`:
+        -   Hints emitted: `minimal` -- the *essential* pointers only,
+            i.e. those without which the just-produced result cannot be
+            consumed at all (such as the follow-up command that turns
+            persisted findings into actual changes)
+    -   If <ase-guidance-level/> is `normal`:
+        -   Hints emitted: `minimal` *and* `normal` -- additionally the
+            *canonical next step* of the finished skill
+    -   If <ase-guidance-level/> is `verbose`:
+        -   Hints emitted: `minimal`, `normal`, *and* `verbose` --
+            additionally the *unused options* of the current skill, the
+            *related* skills, and the *configuration* skills
+    -   If <ase-guidance-level/> is empty or not set:
+        -   Treat it as `normal`
+
+-   *IMPORTANT*: A hint *MUST* be emitted *exclusively* through the
+    `<ase-tpl-hint/>` template pattern below, and *never* as free-text
+    prose, so that its suppression is uniform across all skills.
+
 Template Patterns
 -----------------
 
@@ -347,6 +409,32 @@ Template Patterns
 
     </template>
 
+-   When `<ase-tpl-hint level="<level/>"><content/></ase-tpl-hint>`
+    should be expanded, first determine whether the hint passes the
+    configured guidance level (see `Guidance Hint Level` above):
+
+    -   Set <emit>false</emit>.
+    -   If <ase-guidance-level/> is `minimal` and <level/> is `minimal`,
+        set <emit>true</emit>.
+    -   If <ase-guidance-level/> is `normal` and <level/> is `minimal`
+        or `normal`, set <emit>true</emit>.
+    -   If <ase-guidance-level/> is `verbose` and <level/> is `minimal`,
+        `normal`, or `verbose`, set <emit>true</emit>.
+    -   If <ase-guidance-level/> is empty or not set, treat it as
+        `normal` and apply the corresponding decision above.
+
+    Then dispatch on <emit/>:
+
+    <if condition="<emit/> is `true`">
+        <template>
+        ⧉ **ASE**: ▷ hint [<level/>]: **<content/>**
+        </template>
+    </if>
+    <else>
+        This construct is expanded into *nothing*.
+        Do not output anything.
+    </else>
+
 -   When `<ase-tpl-boxline><line/></ase-tpl-boxline>` should be expanded, use:
 
     <if condition="<line/> is not empty">

package/plugin/package.json

@@ -6,7 +6,7 @@
     "homepage":                             "https://ase.tools",
     "repository":                           { "url": "git+https://github.com/rse/ase.git", "type": "git" },
     "bugs":                                 { "url": "https://github.com/rse/ase/issues" },
-    "version":                              "0.9.47",
+    "version":                              "0.9.49",
     "license":                              "Apache-2.0",
     "author": {
         "name":                             "Dr. Ralf S. Engelschall",

package/plugin/skills/ase-arch-analyze/SKILL.md

@@ -465,14 +465,16 @@ interface quality, quality attributes, and architecture governance.
       "<title/>: <description/>" }` entry per reported PROBLEM and one
       `{ command: "set", key: "ase-issue-T<n/>", val: "<title/>:
       <description/>" }` entry per reported TRADEOFF.
-    </step>
 
-4.  <step id="STEP 4: Give Final Hint">
-    Finally, output the following <template/> to give a final hint:
+    Finally, give a final hint by expanding the following (which,
+    depending on the configured <ase-guidance-level/>, may expand into
+    nothing and hence emit no output at all):
+
+    <ase-tpl-hint level="minimal">
+    For deeper analysis, suggestions on solution approaches and then final source code changes, use `/ase-code-resolve P{n}` or `/ase-code-resolve T{n}` in the same or even a different session.
+    </ase-tpl-hint>
 
-    <template>
-    ⧉ **ASE**: ↪ hint: **For deeper analysis, suggestions on solution approaches and then final source code changes, use `/ase-code-resolve P{n}` or `/ase-code-resolve T{n}` in the same or even a different session.**
-    </template>
     </step>
+
 </flow>
 

package/plugin/skills/ase-arch-discover/SKILL.md

@@ -238,6 +238,18 @@ for the technology stack to *provide* the *needed functionality*
         [...]
         | **<name-N/>** | `<package-N/>` | <version-N/> | **<downloads-N/>** | **<stars-N/>** | **<updated-N/>** | <created-N/> | <deps-N/>  |
         </template>
+
+    3.  Finally, give the closing hints by expanding the following
+        (which, depending on the configured <ase-guidance-level/>, may
+        each expand into nothing and hence emit no output at all):
+
+        <ase-tpl-hint level="normal">
+        To pick a single component from the ranking against weighted criteria, use `/ase-meta-evaluate`.
+        </ase-tpl-hint>
+
+        <ase-tpl-hint level="verbose">
+        Use `/ase-arch-discover --limit` to widen the candidate set, `--staleness` to change the aging threshold, and `--small-scope` to prefer small, focused components.
+        </ase-tpl-hint>
     </step>
 </flow>
 

package/plugin/skills/ase-code-analyze/SKILL.md

@@ -196,11 +196,13 @@ problems in *performance* and *efficiency*, or problems in *security*.
         followed by one `{ command: "set", key: "ase-issue-P<n/>", val:
         "<title/>: <description/>" }` entry per reported problem.
 
-    Finally, output the following <template/> to give a final hint:
+    Finally, give a final hint by expanding the following (which,
+    depending on the configured <ase-guidance-level/>, may expand into
+    nothing and hence emit no output at all):
 
-    <template>
-    ⧉ **ASE**: ↪ hint: **For deeper analysis, suggestions on solution approaches and then final problem resolution, use `/ase-code-resolve P{n}` in the same or even a different session.**
-    </template>
+    <ase-tpl-hint level="minimal">
+    For deeper analysis, suggestions on solution approaches and then final problem resolution, use `/ase-code-resolve P{n}` in the same or even a different session.
+    </ase-tpl-hint>
 
     You *MUST* not output anything else in this STEP 3,
     especially not any further explanations.

package/plugin/skills/ase-code-lint/SKILL.md

@@ -431,6 +431,22 @@ related to a set of code quality aspects.
         <ase-tpl-bullet-secondary/> **LINT FINISHED**
         </template>
 
+    5.  Finally, give the closing hints by expanding the following
+        (which, depending on the configured <ase-guidance-level/>, may
+        each expand into nothing and hence emit no output at all):
+
+        <if condition="at least one problem in <problems/> was left uncorrected">
+        <ase-tpl-hint level="normal">
+        For problems whose correction demands more than a local edit, use `/ase-code-resolve` to derive and apply a full solution approach.
+        </ase-tpl-hint>
+        </if>
+
+        <if condition="<getopt-option-auto/> is not equal `true`">
+        <ase-tpl-hint level="verbose">
+        Use `/ase-code-lint --auto` to apply all corrections unattended, and `--severity` to raise the reporting floor.
+        </ase-tpl-hint>
+        </if>
+
     </step>
 
 </flow>

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

@@ -322,6 +322,16 @@ Analyze documents for spelling, punctuation, or grammar errors
         <ase-tpl-bullet-secondary/> **PROOFREAD FINISHED**
         </template>
 
+    5.  Finally, give the closing hint by expanding the following
+        (which, depending on the configured <ase-guidance-level/>, may
+        expand into nothing and hence emit no output at all):
+
+        <if condition="<getopt-option-auto/> is not equal `true`">
+        <ase-tpl-hint level="verbose">
+        Use `/ase-docs-proofread --auto` to apply all corrections unattended.
+        </ase-tpl-hint>
+        </if>
+
     </step>
 
 </flow>

package/plugin/skills/ase-help-intent/SKILL.md

@@ -102,16 +102,11 @@ catalog you match <intent/> against:
         <template>
         <ase-tpl-head title="SKILL COMMAND PROPOSAL"/>
 
-        ●   **INTENT**:
-        ○   <intent/>
-
-        ●   **COMMAND**:
-        ⌘   `<command/>`
-
-        ●   **RATIONALE**:
-        ○   <rationale/>
+        ❯ `<command/>`
 
         <ase-tpl-foot title="SKILL COMMAND PROPOSAL"/>
+
+        **RATIONALE**: <rationale/>
         </template>
 
     4.  *Dispatch Command*:

package/plugin/skills/ase-help-skill/SKILL.md

@@ -63,9 +63,17 @@ the *sole* index <skill-ref/> is resolved against:
         <catalog/>
 
         <ase-tpl-foot title="SKILL CATALOG"/>
-
-        ⧉ **ASE**: ✪ skill: **ase-help-skill**, ▶ hint: **run `/ase-help-skill ase-xxx-xxx` for manual page of individual skill**
         </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

package/plugin/skills/ase-help-skill/catalog.md

@@ -55,6 +55,6 @@
 ○   `ase-task-delete`:     Delete a Task Plan
 
 ⎈   **OTHER SKILLS**
-○   `ase-meta-persona`:    Persona Configuration
+○   `ase-meta-config`:     Configuration Management
 ○   `ase-meta-compat`:     Self-Test ASE Compatibility
 

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

@@ -271,6 +271,14 @@ Honor the following tenets throughout the brainstorming:
         <ase-tpl-bullet-signal/> **RECOMMENDED IDEA**: <recommendation/>
         </template>
 
+    5.  Finally, give the closing hint by expanding the following (which,
+        depending on the configured <ase-guidance-level/>, may expand
+        into nothing and hence emit no output at all):
+
+        <ase-tpl-hint level="normal">
+        To turn the recommended idea into a plan, use `/ase-task-edit`; to turn it directly into code, use `/ase-code-craft`.
+        </ase-tpl-hint>
+
     </step>
 
 </flow>

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

@@ -71,6 +71,19 @@ currently staged Git changes.
 
     Do *not* output any further explanation.
 
+    Finally, give the closing hint by expanding the following (which,
+    depending on the configured <ase-guidance-level/>, may expand into
+    nothing and hence emit no output at all):
+
+    <if condition="
+        a `CHANGELOG.md` file exists in the project (or in any affected
+        sub-package) and is *not* itself part of the staged changes
+    ">
+    <ase-tpl-hint level="normal">
+    The staged changes do not touch `CHANGELOG.md` -- use `/ase-meta-changelog` to update its entries before committing.
+    </ase-tpl-hint>
+    </if>
+
     </step>
 
 </flow>

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

@@ -0,0 +1,165 @@
+---
+name: ase-meta-config
+argument-hint: "[--help|-h] [--scope|-s <scope>] <operation> [<args>]"
+description: >
+    List, get, set, or delete the layered ASE configuration values across
+    the user/project/task/session scope chain.
+    Use when the user wants to "configure" ASE, or to inspect or change a
+    configuration key like `agent.persona`, `agent.guidance`, `agent.task`,
+    or `project.boxing`.
+user-invocable: true
+disable-model-invocation: false
+effort: high
+---
+
+@${CLAUDE_SKILL_DIR}/../../meta/ase-control.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md
+@${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md
+
+<skill name="ase-meta-config">
+Configuration Management
+</skill>
+
+<expand name="getopt"
+    arg1="ase-meta-config"
+    arg2="--scope|-s=">
+    $ARGUMENTS
+</expand>
+
+<objective>
+*List*, *get*, *set*, or *delete* the values of the *layered
+configuration* of ASE, mirroring the non-interactive `ase config
+<operation>` CLI subcommands *exclusively* through the `ase` MCP server.
+</objective>
+
+Procedure
+---------
+
+1.  **Determine Operation:**
+
+    1.  Set <arguments><getopt-arguments/></arguments>, with any leading
+        and trailing whitespace stripped. Split <arguments/> into
+        whitespace-separated tokens, honoring single and double quotes
+        (a quoted token is *one* token, with its surrounding quotes
+        removed). Set <operation/> to the *first* token, lower-cased,
+        and <operands/> to the list of the *remaining* tokens. If
+        <arguments/> is empty, set <operation>(none)</operation> and
+        <operands/> to the empty list. Inherit the always existing
+        <ase-session-id/> from the current context.
+        Do not output anything.
+
+    2.  Determine the target scope chain:
+
+        <if condition="<getopt-option-scope/> is empty">
+        Set <scope>session:<ase-session-id/></scope>. Reads then cascade
+        `user` -> `project` -> `session`, and writes land on the *session*
+        layer -- the only layer on which the `agent.task` and `agent.skill`
+        keys are writable at all.
+        </if>
+        <else>
+        Set <scope><getopt-option-scope/></scope>, forwarding the given
+        scope chain verbatim.
+        </else>
+
+        Do not output anything.
+
+    3.  You *MUST* *NOT* use `Bash`, `Read`, `Write`, `Edit`, or any
+        other filesystem-touching tool anywhere in this skill. *Every*
+        configuration access is performed *exclusively* through the
+        `ase_config_*` tools of the `ase` MCP server.
+        Do not output anything.
+
+2.  **Dispatch Operation:**
+
+    1.  <if condition="<operation/> is `list` AND <operands/> is empty">
+        Call the `ase_config_list(scope: "<scope/>")` tool from the `ase`
+        MCP server. The result is a structured object with an `entries`
+        array where each entry has a `key`, a `value`, and a `scope` field.
+
+        -   If the `entries` array is empty, output the following <template/>:
+
+            <template>
+            ⧉ **ASE**: ⚙ config (scope: `<scope/>`): *(none)*
+            </template>
+
+        -   Else output the following <template/>, where each <key/>,
+            <value/>, and <entry-scope/> correspond to one entry of the
+            `entries` array, in the given order:
+
+            <template>
+            ⧉ **ASE**: ⚙ config (scope: `<scope/>`):
+
+            | *Key*            | *Value*          | *Scope*          |
+            |------------------|------------------|------------------|
+            | `<key/>`         | **<value/>**     | `<entry-scope/>` |
+            | [...]            | [...]            | [...]            |
+
+            </template>
+        </if>
+
+    2.  <elseif condition="<operation/> is `get` AND <operands/> has exactly one token">
+        Set <key/> to the single token of <operands/>. Call the
+        `ase_config_get(key: "<key/>", scope: "<scope/>")` tool from the
+        `ase` MCP server and set <text/> to its `text` output field.
+
+        -   If <text/> is empty, the key is not set at all. Output the
+            following <template/>:
+
+            <template>
+            ⧉ **ASE**: ⚙ config: `<key/>` (scope: `<scope/>`): *(not set)*
+            </template>
+
+        -   Else set <value/> to the JSON-decoded <text/> and output the
+            following <template/>:
+
+            <template>
+            ⧉ **ASE**: ⚙ config: `<key/>` (scope: `<scope/>`): **<value/>**
+            </template>
+        </elseif>
+
+    3.  <elseif condition="<operation/> is `set` AND <operands/> has exactly two tokens">
+        Set <key/> to the *first* and <value/> to the *second* token of
+        <operands/>. Call the `ase_config_set(key: "<key/>", val:
+        "<value/>", scope: "<scope/>")` tool from the `ase` MCP server.
+
+        Then, if <key/> is one of the three keys steering your *own*
+        behaviour, you *MUST* immediately adopt the new <value/> for
+        the remainder of the session -- *including* the output of this
+        very skill run -- by re-evaluating and internalizing the
+        corresponding rules of the constitution: for `agent.persona` set
+        <ase-persona-style><value/></ase-persona-style> and re-evaluate
+        the `Persona Communication Style` rules, for `agent.guidance` set
+        <ase-guidance-level><value/></ase-guidance-level> and re-evaluate
+        the `Guidance Hint Level` rules, and for `project.boxing` set
+        <ase-project-boxing><value/></ase-project-boxing> and re-evaluate
+        the `Artifact Boxing Transparency` rules. Do not output anything
+        for this.
+
+        Then only output the following <template/>:
+
+        <template>
+        ⧉ **ASE**: ⚙ config: `<key/>` (scope: `<scope/>`): **<value/>** (*updated*)
+        </template>
+        </elseif>
+
+    4.  <elseif condition="<operation/> is `delete` AND <operands/> has exactly one token">
+        Set <key/> to the single token of <operands/>. Call the
+        `ase_config_delete(key: "<key/>", scope: "<scope/>")` tool from
+        the `ase` MCP server. Then only output the following <template/>:
+
+        <template>
+        ⧉ **ASE**: ⚙ config: `<key/>` (scope: `<scope/>`): (*deleted*)
+        </template>
+        </elseif>
+
+    5.  <else>
+        The <operation/> is either missing, unknown, or was given with
+        the wrong number of operands. Notice that the `init` and `edit`
+        subcommands of the `ase config` CLI are deliberately *not*
+        mirrored by this skill. Only output the following <template/>
+        and then immediately *STOP* processing the entire current skill:
+
+        <template>
+        ⧉ **ASE**: ☻ skill: **ase-meta-config**, ▶ ERROR: invalid operation: **<operation/>** (expected `list`, `get <key>`, `set <key> <value>`, or `delete <key>`)
+        </template>
+        </else>

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

@@ -0,0 +1,136 @@
+
+##  NAME
+
+`ase-meta-config` - Configuration Management
+
+##  SYNOPSIS
+
+`ase-meta-config`
+    [`--help`|`-h`]
+    [`--scope`|`-s` *scope*]
+    *operation*
+    [*args*]
+
+##  DESCRIPTION
+
+The `ase-meta-config` skill lists, reads, writes, and removes the values
+of the *layered ASE configuration* from inside the assistant session. It
+mirrors the non-interactive subcommands of the `ase config` CLI, but
+performs *every* access through the `ase_config_list`, `ase_config_get`,
+`ase_config_set`, and `ase_config_delete` tools of the `ase` MCP server -
+it never shells out and never touches a file directly.
+
+The configuration is organized as a chain of *scopes*, canonically
+ordered `default` < `user` < `project` < `task` < `session`. Reads
+cascade from the strongest (rightmost) scope down to the weakest and
+return the first value that is defined; writes are always confined to
+the strongest (target) scope of the chain.
+
+The recognized keys are grouped under `project.*` (`project.id`,
+`project.name`, `project.boxing`, and the
+`project.artifact.`*kind*`.{basedir,files}` globs) and `agent.*`
+(`agent.persona`, `agent.guidance`, `agent.task`, and `agent.skill`).
+Some keys are writable on selected scopes only; in particular
+`agent.task` and `agent.skill` are writable on a `session` scope only,
+and the `project.artifact.*` globs are writable on the `user` and
+`project` scopes only.
+
+The three keys steering the behaviour of the agent itself accept fixed
+value sets: `agent.persona` selects the *communication style* and is one
+of the decorative, eloquent, and explaining `writer`, the concise,
+factual, and accurate `engineer` (the default), the layered,
+pyramid-structured `journalist`, the brief, factual, and abbreviating
+`telegrapher`, or the terse, rough, and stuttering `caveman`;
+`agent.guidance` selects the amount of unsolicited hints and is one of
+`none`, `minimal`, `normal` (the default), or `verbose`; and
+`project.boxing` selects the artifact transparency and is one of `white`,
+`grey`, or `black`. Setting `agent.persona`, `agent.guidance`, or
+`project.boxing` takes effect immediately, i.e. still within the running
+session.
+
+The following *operations* exist:
+
+- `list`: List all effective configuration entries of the scope chain as
+  a table of key, value, and the scope that supplied the value. For
+  overlapping keys only the value of the strongest scope is shown.
+
+- `get` *key*: Report the effective value at the dotted *key*, or
+  `(not set)` if no scope of the chain defines it.
+
+- `set` *key* *value*: Write *value* at the dotted *key* on the target
+  scope. The value is validated against the configuration schema before
+  it is persisted.
+
+- `delete` *key*: Remove the value at the dotted *key* from the target
+  scope. A key that is not present is silently ignored.
+
+The `init` and `edit` subcommands of the `ase config` CLI are
+deliberately *not* mirrored: `edit` is bound to the interactive `$EDITOR`
+and therefore has no meaning inside an assistant turn, and `init` is a
+preset-bootstrapping operation that stays a shell concern next to `ase
+setup`. Both remain available as `ase config init` and `ase config edit`
+on the command line.
+
+##  OPTIONS
+
+`--scope`|`-s` *scope*:
+    The scope chain to operate on, given as a comma-separated list of
+    `user`, `project`, `task:`*id*, and/or `session:`*id* terms, in any
+    order and at most one term per kind. The `user` term is always
+    implicitly added at the bottom of the chain, and `project` is
+    implicitly added whenever a project context exists. If the option is
+    omitted, the chain of the *current session* (`session:`*id*) is used,
+    so that reads see the full `user` -> `project` -> `session` cascade
+    and writes -- including those to the session-only keys `agent.task`
+    and `agent.skill` -- land on the session layer.
+
+##  ARGUMENTS
+
+*operation*:
+    The operation to perform; one of `list`, `get`, `set`, or `delete`.
+
+*args*:
+    The operands of the operation: none for `list`, a dotted *key* for
+    `get` and `delete`, and a dotted *key* plus a *value* for `set`.
+
+##  EXAMPLES
+
+List all effective configuration entries:
+
+```text
+❯ /ase-meta-config list
+```
+
+Report the effective persona style:
+
+```text
+❯ /ase-meta-config get agent.persona
+```
+
+Switch the communication style for the current session only:
+
+```text
+❯ /ase-meta-config set agent.persona telegrapher
+```
+
+Switch the project boxing for the current session only:
+
+```text
+❯ /ase-meta-config set project.boxing grey
+```
+
+Persist the guidance level for the whole project instead:
+
+```text
+❯ /ase-meta-config --scope project set agent.guidance minimal
+```
+
+Remove a value from the user-level configuration:
+
+```text
+❯ /ase-meta-config --scope user delete project.name
+```
+
+##  SEE ALSO
+
+[`ase-task-id`](../ase-task-id/help.md).

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

@@ -74,5 +74,5 @@ Surface at least fifteen anti-theses:
 ##  SEE ALSO
 
 [`ase-meta-why`](../ase-meta-why/help.md), [`ase-meta-evaluate`](../ase-meta-evaluate/help.md), [`ase-meta-quorum`](../ase-meta-quorum/help.md),
-[`ase-meta-persona`](../ase-meta-persona/help.md).
+[`ase-meta-steelman`](../ase-meta-steelman/help.md).