syntaur 0.17.0 → 0.17.2

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "syntaur",
-  "version": "0.17.0",
+  "version": "0.17.2",
   "description": "Project workflow CLI with dashboard, Claude Code plugin, and Codex plugin",
   "homepage": "https://github.com/prong-horn/syntaur#readme",
   "repository": {

package/platforms/claude-code/.claude-plugin/plugin.json

@@ -29,6 +29,7 @@
     "./skills/claim-resource",
     "./skills/release-resource",
     "./skills/extend-resource",
-    "./skills/list-resources"
+    "./skills/list-resources",
+    "./skills/run-playbook"
   ]
 }

package/platforms/claude-code/commands/run-playbook/run-playbook.md

@@ -0,0 +1,16 @@
+---
+name: run-playbook
+description: Load a Syntaur playbook's full content and follow its directives (resolves by name/slug against the manifest)
+arguments:
+  - name: args
+    description: "<playbook name or slug> (omit to list available playbooks)"
+    required: false
+---
+
+# /run-playbook
+
+Thin wrapper that invokes the `run-playbook` skill via the Skill tool. The skill resolves a playbook by name/slug against `~/.syntaur/playbooks/manifest.md`, loads its full content, and follows its directives. With no argument it lists the available playbooks.
+
+Arguments: $ARGUMENTS
+
+If the skill is not installed, tell the user to run `syntaur install-plugin`.

package/platforms/claude-code/skills/run-playbook/SKILL.md

@@ -0,0 +1,87 @@
+---
+name: run-playbook
+description: >-
+  Load a Syntaur playbook's full content on demand and follow its directives for
+  the rest of the relevant work. Use when the user wants to "run the X playbook",
+  "apply the X playbook", "follow the X playbook", or "load the X playbook" — or
+  asks which playbooks are available. Playbooks live at `~/.syntaur/playbooks/`
+  and define behavioral rules for agents; this skill resolves one by name/slug
+  against the manifest and applies it. Read-only — never writes playbook files.
+license: MIT
+metadata:
+  author: prong-horn
+  version: "1.0.0"
+---
+
+# Run Playbook
+
+Resolve a Syntaur playbook by name or slug, load its full content, and follow
+its directives. Playbooks are behavioral rules for AI agents stored at
+`~/.syntaur/playbooks/<slug>.md` and indexed by the derived
+`~/.syntaur/playbooks/manifest.md`. They are normally only auto-injected as a
+one-line summary via the UserPromptSubmit hook; this skill loads the **full**
+text of one playbook so you can deliberately apply it mid-session.
+
+## When NOT to use this skill
+
+- The user wants to **create, edit, enable, disable, or delete** a playbook —
+  use the `syntaur create-playbook` / `enable-playbook` / `disable-playbook` /
+  `delete-playbook` CLI commands instead. This skill is read-only and must
+  **NEVER** write `manifest.md` or any playbook file.
+- The user just wants the one-line summary of all playbooks — that is already
+  auto-injected each turn; only run this skill when they want the full content
+  applied.
+
+## Step 1: Parse the argument
+
+Take the playbook name or slug from the argument (e.g. `commit-discipline`,
+`Commit Discipline`, `commit`).
+
+If the argument is **empty or ambiguous** (could match more than one playbook),
+do not guess. Read `~/.syntaur/playbooks/manifest.md` and present every
+available playbook — Name, slug, the one-line description, and the
+`_When to use_` line — then ask the user which one to run. Stop until they pick.
+
+## Step 2: Resolve the playbook
+
+Read `~/.syntaur/playbooks/manifest.md`. It is the **enabled-only** index
+(disabled playbooks are excluded), so it is the correct list to resolve against.
+Each entry looks like:
+
+```markdown
+- **[Commit Discipline](commit-discipline.md)** — Make small, logical commits …
+  _When to use: When making git commits during assignment work_
+```
+
+The slug is the link target stem (`commit-discipline` from `commit-discipline.md`).
+Match the user's argument **case-insensitively** against:
+
+1. the slug (link target stem),
+2. the display Name, or
+3. the filename.
+
+If nothing matches, suggest the closest manifest entries (by name/slug
+similarity) and stop — **never invent a slug** or read a file that is not in
+the manifest.
+
+## Step 3: Load the playbook
+
+Read the full `~/.syntaur/playbooks/<slug>.md` file. Note its frontmatter
+(`name`, `slug`, `description`, `when_to_use`, `tags`) and read the **entire**
+markdown body — that body is the set of directives.
+
+## Step 4: Apply the playbook
+
+Treat the playbook body as behavioral directives that **take precedence over
+default conventions** for the remainder of the relevant work (consistent with
+how playbooks are described in the `syntaur-protocol` skill). Keep following
+them for the rest of the session's work on this topic, not just the next action.
+
+## Step 5: Report to User
+
+Confirm:
+
+- Which playbook was loaded — display name, slug, and absolute path.
+- A one-line summary of what it now governs (from its `when_to_use` / first
+  directives), e.g. "Now applying **Commit Discipline**: small, logical commits
+  tied to plan tasks."

package/platforms/codex/skills/run-playbook/SKILL.md

@@ -0,0 +1,87 @@
+---
+name: run-playbook
+description: >-
+  Load a Syntaur playbook's full content on demand and follow its directives for
+  the rest of the relevant work. Use when the user wants to "run the X playbook",
+  "apply the X playbook", "follow the X playbook", or "load the X playbook" — or
+  asks which playbooks are available. Playbooks live at `~/.syntaur/playbooks/`
+  and define behavioral rules for agents; this skill resolves one by name/slug
+  against the manifest and applies it. Read-only — never writes playbook files.
+license: MIT
+metadata:
+  author: prong-horn
+  version: "1.0.0"
+---
+
+# Run Playbook
+
+Resolve a Syntaur playbook by name or slug, load its full content, and follow
+its directives. Playbooks are behavioral rules for AI agents stored at
+`~/.syntaur/playbooks/<slug>.md` and indexed by the derived
+`~/.syntaur/playbooks/manifest.md`. They are normally only auto-injected as a
+one-line summary via the UserPromptSubmit hook; this skill loads the **full**
+text of one playbook so you can deliberately apply it mid-session.
+
+## When NOT to use this skill
+
+- The user wants to **create, edit, enable, disable, or delete** a playbook —
+  use the `syntaur create-playbook` / `enable-playbook` / `disable-playbook` /
+  `delete-playbook` CLI commands instead. This skill is read-only and must
+  **NEVER** write `manifest.md` or any playbook file.
+- The user just wants the one-line summary of all playbooks — that is already
+  auto-injected each turn; only run this skill when they want the full content
+  applied.
+
+## Step 1: Parse the argument
+
+Take the playbook name or slug from the argument (e.g. `commit-discipline`,
+`Commit Discipline`, `commit`).
+
+If the argument is **empty or ambiguous** (could match more than one playbook),
+do not guess. Read `~/.syntaur/playbooks/manifest.md` and present every
+available playbook — Name, slug, the one-line description, and the
+`_When to use_` line — then ask the user which one to run. Stop until they pick.
+
+## Step 2: Resolve the playbook
+
+Read `~/.syntaur/playbooks/manifest.md`. It is the **enabled-only** index
+(disabled playbooks are excluded), so it is the correct list to resolve against.
+Each entry looks like:
+
+```markdown
+- **[Commit Discipline](commit-discipline.md)** — Make small, logical commits …
+  _When to use: When making git commits during assignment work_
+```
+
+The slug is the link target stem (`commit-discipline` from `commit-discipline.md`).
+Match the user's argument **case-insensitively** against:
+
+1. the slug (link target stem),
+2. the display Name, or
+3. the filename.
+
+If nothing matches, suggest the closest manifest entries (by name/slug
+similarity) and stop — **never invent a slug** or read a file that is not in
+the manifest.
+
+## Step 3: Load the playbook
+
+Read the full `~/.syntaur/playbooks/<slug>.md` file. Note its frontmatter
+(`name`, `slug`, `description`, `when_to_use`, `tags`) and read the **entire**
+markdown body — that body is the set of directives.
+
+## Step 4: Apply the playbook
+
+Treat the playbook body as behavioral directives that **take precedence over
+default conventions** for the remainder of the relevant work (consistent with
+how playbooks are described in the `syntaur-protocol` skill). Keep following
+them for the rest of the session's work on this topic, not just the next action.
+
+## Step 5: Report to User
+
+Confirm:
+
+- Which playbook was loaded — display name, slug, and absolute path.
+- A one-line summary of what it now governs (from its `when_to_use` / first
+  directives), e.g. "Now applying **Commit Discipline**: small, logical commits
+  tied to plan tasks."

package/skills/run-playbook/SKILL.md

@@ -0,0 +1,87 @@
+---
+name: run-playbook
+description: >-
+  Load a Syntaur playbook's full content on demand and follow its directives for
+  the rest of the relevant work. Use when the user wants to "run the X playbook",
+  "apply the X playbook", "follow the X playbook", or "load the X playbook" — or
+  asks which playbooks are available. Playbooks live at `~/.syntaur/playbooks/`
+  and define behavioral rules for agents; this skill resolves one by name/slug
+  against the manifest and applies it. Read-only — never writes playbook files.
+license: MIT
+metadata:
+  author: prong-horn
+  version: "1.0.0"
+---
+
+# Run Playbook
+
+Resolve a Syntaur playbook by name or slug, load its full content, and follow
+its directives. Playbooks are behavioral rules for AI agents stored at
+`~/.syntaur/playbooks/<slug>.md` and indexed by the derived
+`~/.syntaur/playbooks/manifest.md`. They are normally only auto-injected as a
+one-line summary via the UserPromptSubmit hook; this skill loads the **full**
+text of one playbook so you can deliberately apply it mid-session.
+
+## When NOT to use this skill
+
+- The user wants to **create, edit, enable, disable, or delete** a playbook —
+  use the `syntaur create-playbook` / `enable-playbook` / `disable-playbook` /
+  `delete-playbook` CLI commands instead. This skill is read-only and must
+  **NEVER** write `manifest.md` or any playbook file.
+- The user just wants the one-line summary of all playbooks — that is already
+  auto-injected each turn; only run this skill when they want the full content
+  applied.
+
+## Step 1: Parse the argument
+
+Take the playbook name or slug from the argument (e.g. `commit-discipline`,
+`Commit Discipline`, `commit`).
+
+If the argument is **empty or ambiguous** (could match more than one playbook),
+do not guess. Read `~/.syntaur/playbooks/manifest.md` and present every
+available playbook — Name, slug, the one-line description, and the
+`_When to use_` line — then ask the user which one to run. Stop until they pick.
+
+## Step 2: Resolve the playbook
+
+Read `~/.syntaur/playbooks/manifest.md`. It is the **enabled-only** index
+(disabled playbooks are excluded), so it is the correct list to resolve against.
+Each entry looks like:
+
+```markdown
+- **[Commit Discipline](commit-discipline.md)** — Make small, logical commits …
+  _When to use: When making git commits during assignment work_
+```
+
+The slug is the link target stem (`commit-discipline` from `commit-discipline.md`).
+Match the user's argument **case-insensitively** against:
+
+1. the slug (link target stem),
+2. the display Name, or
+3. the filename.
+
+If nothing matches, suggest the closest manifest entries (by name/slug
+similarity) and stop — **never invent a slug** or read a file that is not in
+the manifest.
+
+## Step 3: Load the playbook
+
+Read the full `~/.syntaur/playbooks/<slug>.md` file. Note its frontmatter
+(`name`, `slug`, `description`, `when_to_use`, `tags`) and read the **entire**
+markdown body — that body is the set of directives.
+
+## Step 4: Apply the playbook
+
+Treat the playbook body as behavioral directives that **take precedence over
+default conventions** for the remainder of the relevant work (consistent with
+how playbooks are described in the `syntaur-protocol` skill). Keep following
+them for the rest of the session's work on this topic, not just the next action.
+
+## Step 5: Report to User
+
+Confirm:
+
+- Which playbook was loaded — display name, slug, and absolute path.
+- A one-line summary of what it now governs (from its `when_to_use` / first
+  directives), e.g. "Now applying **Commit Discipline**: small, logical commits
+  tied to plan tasks."

package/dashboard/dist/assets/channel-eXdj7ayV.js

@@ -1 +0,0 @@
-import{aq as o,ar as n}from"./mermaid.core-D5JD5uYg.js";const t=(r,a)=>o.lang.round(n.parse(r)[a]);export{t as c};

package/dashboard/dist/assets/classDiagram-VBA2DB6C-B-bBQACr.js

@@ -1 +0,0 @@
-import{s as a,c as s,a as e,C as t}from"./chunk-WL4C6EOR-mY_1Vmdf.js";import{_ as i}from"./mermaid.core-D5JD5uYg.js";import"./chunk-FMBD7UC4-BVPaoRtV.js";import"./chunk-JSJVCQXG-B9uJpMzd.js";import"./chunk-55IACEB6-2vCDm2cJ.js";import"./chunk-KX2RTZJC-CHcjjGaq.js";import"./index--qwybK3W.js";var n={parser:e,get db(){return new t},renderer:s,styles:a,init:i(r=>{r.class||(r.class={}),r.class.arrowMarkerAbsolute=r.arrowMarkerAbsolute},"init")};export{n as diagram};

package/dashboard/dist/assets/classDiagram-v2-RAHNMMFH-B-bBQACr.js

@@ -1 +0,0 @@
-import{s as a,c as s,a as e,C as t}from"./chunk-WL4C6EOR-mY_1Vmdf.js";import{_ as i}from"./mermaid.core-D5JD5uYg.js";import"./chunk-FMBD7UC4-BVPaoRtV.js";import"./chunk-JSJVCQXG-B9uJpMzd.js";import"./chunk-55IACEB6-2vCDm2cJ.js";import"./chunk-KX2RTZJC-CHcjjGaq.js";import"./index--qwybK3W.js";var n={parser:e,get db(){return new t},renderer:s,styles:a,init:i(r=>{r.class||(r.class={}),r.class.arrowMarkerAbsolute=r.arrowMarkerAbsolute},"init")};export{n as diagram};

package/dashboard/dist/assets/clone-CL04pv9J.js

@@ -1 +0,0 @@
-import{b as r}from"./_baseUniq-BeLddDKU.js";var e=4;function a(o){return r(o,e)}export{a as c};