onbuzz 4.8.0 → 4.8.1

package/src/tools/skillsTool.js

@@ -1,277 +1,282 @@
-/**
- * Skills Tool - Global skills library for agents
- *
- * Purpose:
- * - Allow agents to discover, browse, and read reusable skill instructions
- * - Support progressive disclosure: list → describe → read-section → read
- * - Support CRUD operations and importing skills from disk
- * - Skills are global (shared across agents) and persist across package updates
- *
- * Actions:
- * - list: List all skills with descriptions, section headings, and sizes
- * - describe: Get full metadata for a skill without loading content
- * - read: Read a skill's full content
- * - read-section: Read only a specific section of a skill
- * - read-file: Read a supporting file from a skill directory
- * - create: Create a new skill
- * - update: Update an existing skill
- * - delete: Remove a skill
- * - import: Import a skill from an external file or directory
- */
-
-import { BaseTool } from './baseTool.js';
-import { getSkillsService } from '../services/skillsService.js';
-import { SKILLS_ACTIONS } from '../utilities/toolConstants.js';
-
-class SkillsTool extends BaseTool {
-  constructor(config = {}, logger = null) {
-    super(config, logger);
-
-    this.skillsService = null;
-    this.requiresProject = false;
-    this.isAsync = false;
-    this.timeout = 30000;
-  }
-
-  async _ensureSkillsService() {
-    if (!this.skillsService) {
-      this.skillsService = getSkillsService(this.logger);
-      await this.skillsService.initialize();
-    }
-    return this.skillsService;
-  }
-
-  getDescription() {
-    return `Skills Tool: Browse and manage a global library of reusable skill instructions.
-
-Skills are structured knowledge packages containing instructions, checklists, templates, and reference files.
-Each skill is a directory with a skill.md file and optional supporting files.
-Always check for relevant skills when tackeling new tasks/changing task-focus.
-
-PROGRESSIVE DISCLOSURE — Use this flow to minimize context usage:
-1. "list" → See all skills with section headings and sizes
-2. "describe" → Inspect a specific skill's structure in detail
-3. "read-section" → Load only the section you need
-4. "read" → Load full content only when necessary
-
-ACTIONS:
-
-1. LIST all skills:
-\`\`\`json
-{ "toolId": "skills", "action": "list" }
-\`\`\`
-Returns: name, description, section headings, line count, file count for each skill.
-
-2. DESCRIBE a skill (metadata only, no content):
-\`\`\`json
-{ "toolId": "skills", "action": "describe", "name": "code-review" }
-\`\`\`
-Returns: description, sections with line ranges, file list, size.
-
-3. READ full skill content:
-\`\`\`json
-{ "toolId": "skills", "action": "read", "name": "code-review" }
-\`\`\`
-Returns: full skill.md content + list of files in the skill directory.
-
-4. READ a specific SECTION:
-\`\`\`json
-{ "toolId": "skills", "action": "read-section", "name": "code-review", "section": "Checklist" }
-\`\`\`
-Returns: only the content under the specified ## heading.
-
-5. READ a supporting FILE:
-\`\`\`json
-{ "toolId": "skills", "action": "read-file", "name": "email-templates", "file": "templates/welcome.html" }
-\`\`\`
-Returns: content of a specific file within the skill directory.
-
-6. CREATE a new skill:
-\`\`\`json
-{ "toolId": "skills", "action": "create", "name": "my-skill", "content": "# My Skill\\n\\nInstructions here...\\n\\n## Section One\\n..." }
-\`\`\`
-Optional: "files" array of { "path": "relative/path.ext", "content": "..." } for supporting files.
-
-7. UPDATE an existing skill:
-\`\`\`json
-{ "toolId": "skills", "action": "update", "name": "my-skill", "content": "# Updated content..." }
-\`\`\`
-Optional: "files" array to add/update supporting files.
-
-8. DELETE a skill:
-\`\`\`json
-{ "toolId": "skills", "action": "delete", "name": "my-skill" }
-\`\`\`
-
-9. IMPORT a skill from disk:
-\`\`\`json
-{ "toolId": "skills", "action": "import", "source": "/path/to/skill-dir-or-file" }
-\`\`\`
-Optional: "name" to override the derived skill name. If source is a directory, it must contain a skill.md file.
-
-SKILL NAMING: Names must be kebab-case (lowercase, hyphens). Example: "code-review", "email-templates".`;
-  }
-
-  parseParameters(content) {
-    return content;
-  }
-
-  getRequiredParameters() {
-    return ['action'];
-  }
-
-  getSupportedActions() {
-    return Object.values(SKILLS_ACTIONS);
-  }
-
-  validateParameterTypes(params) {
-    const errors = [];
-    if (params.action && typeof params.action !== 'string') {
-      errors.push('action must be a string');
-    }
-    if (params.name !== undefined && typeof params.name !== 'string') {
-      errors.push('name must be a string');
-    }
-    if (params.content !== undefined && typeof params.content !== 'string') {
-      errors.push('content must be a string');
-    }
-    if (params.section !== undefined && typeof params.section !== 'string') {
-      errors.push('section must be a string');
-    }
-    if (params.file !== undefined && typeof params.file !== 'string') {
-      errors.push('file must be a string');
-    }
-    if (params.source !== undefined && typeof params.source !== 'string') {
-      errors.push('source must be a string');
-    }
-    return errors;
-  }
-
-  customValidateParameters(params) {
-    const errors = [];
-    const { action, name, content, section, file, source } = params;
-
-    const validActions = this.getSupportedActions();
-    if (!validActions.includes(action)) {
-      errors.push(`Invalid action: "${action}". Valid actions: ${validActions.join(', ')}`);
-      return errors;
-    }
-
-    // Action-specific required params
-    const needsName = [SKILLS_ACTIONS.DESCRIBE, SKILLS_ACTIONS.READ, SKILLS_ACTIONS.READ_SECTION, SKILLS_ACTIONS.READ_FILE, SKILLS_ACTIONS.CREATE, SKILLS_ACTIONS.UPDATE, SKILLS_ACTIONS.DELETE];
-    if (needsName.includes(action) && !name) {
-      errors.push(`"name" is required for action "${action}"`);
-    }
-    if (action === SKILLS_ACTIONS.CREATE && !content) {
-      errors.push('"content" is required for action "create"');
-    }
-    if (action === SKILLS_ACTIONS.READ_SECTION && !section) {
-      errors.push('"section" is required for action "read-section"');
-    }
-    if (action === SKILLS_ACTIONS.READ_FILE && !file) {
-      errors.push('"file" is required for action "read-file"');
-    }
-    if (action === SKILLS_ACTIONS.IMPORT && !source) {
-      errors.push('"source" is required for action "import"');
-    }
-
-    return errors;
-  }
-
-  async execute(params, context = {}) {
-    const service = await this._ensureSkillsService();
-    const { action } = params;
-
-    try {
-      switch (action) {
-        case SKILLS_ACTIONS.LIST:
-          return this._formatResult(await service.listSkills(), 'Skills listed');
-
-        case SKILLS_ACTIONS.DESCRIBE:
-          return this._formatResult(await service.describeSkill(params.name), `Skill described: ${params.name}`);
-
-        case SKILLS_ACTIONS.READ:
-          return this._formatResult(await service.readSkill(params.name), `Skill read: ${params.name}`);
-
-        case SKILLS_ACTIONS.READ_SECTION:
-          return this._formatResult(await service.readSkillSection(params.name, params.section), `Section read: ${params.section}`);
-
-        case SKILLS_ACTIONS.READ_FILE:
-          return this._formatResult(await service.readSkillFile(params.name, params.file), `File read: ${params.file}`);
-
-        case SKILLS_ACTIONS.CREATE:
-          return this._formatResult(await service.createSkill(params.name, params.content, params.files || [], params.description || null), `Skill created: ${params.name}`);
-
-        case SKILLS_ACTIONS.UPDATE:
-          return this._formatResult(await service.updateSkill(params.name, params.content || null, params.files || [], params.description || null), `Skill updated: ${params.name}`);
-
-        case SKILLS_ACTIONS.DELETE:
-          await service.deleteSkill(params.name);
-          return this._formatResult({ deleted: params.name }, `Skill deleted: ${params.name}`);
-
-        case SKILLS_ACTIONS.IMPORT:
-          return this._formatResult(await service.importSkill(params.source, params.name || null, params.description || null), `Skill imported: ${params.name || params.source}`);
-
-        default:
-          return { success: false, error: `Unknown action: ${action}` };
-      }
-    } catch (error) {
-      return { success: false, error: error.message };
-    }
-  }
-
-  _formatResult(data, message) {
-    return {
-      success: true,
-      result: data,
-      message
-    };
-  }
-
-  getParameterSchema() {
-    return {
-      type: 'object',
-      required: ['action'],
-      properties: {
-        action: {
-          type: 'string',
-          enum: Object.values(SKILLS_ACTIONS),
-          description: 'The skill action to perform'
-        },
-        name: {
-          type: 'string',
-          description: 'Skill name (kebab-case)'
-        },
-        content: {
-          type: 'string',
-          description: 'Skill content (markdown)'
-        },
-        section: {
-          type: 'string',
-          description: 'Section heading to read (for read-section action)'
-        },
-        file: {
-          type: 'string',
-          description: 'Relative file path within skill directory (for read-file action)'
-        },
-        source: {
-          type: 'string',
-          description: 'Source file or directory path (for import action)'
-        },
-        files: {
-          type: 'array',
-          items: {
-            type: 'object',
-            properties: {
-              path: { type: 'string' },
-              content: { type: 'string' }
-            }
-          },
-          description: 'Additional supporting files (for create/update actions)'
-        }
-      }
-    };
-  }
-}
-
-export default SkillsTool;
+/**
+ * Skills Tool - Global skills library for agents
+ *
+ * Purpose:
+ * - Allow agents to discover, browse, and read reusable skill instructions
+ * - Support progressive disclosure: list → describe → read-section → read
+ * - Support CRUD operations and importing skills from disk
+ * - Skills are global (shared across agents) and persist across package updates
+ *
+ * Actions:
+ * - list: List all skills with descriptions, section headings, and sizes
+ * - describe: Get full metadata for a skill without loading content
+ * - read: Read a skill's full content
+ * - read-section: Read only a specific section of a skill
+ * - read-file: Read a supporting file from a skill directory
+ * - create: Create a new skill
+ * - update: Update an existing skill
+ * - delete: Remove a skill
+ * - import: Import a skill from an external file or directory
+ */
+
+import { BaseTool } from './baseTool.js';
+import { getSkillsService } from '../services/skillsService.js';
+import { SKILLS_ACTIONS } from '../utilities/toolConstants.js';
+
+class SkillsTool extends BaseTool {
+  constructor(config = {}, logger = null) {
+    super(config, logger);
+
+    this.skillsService = null;
+    this.requiresProject = false;
+    this.isAsync = false;
+    this.timeout = 30000;
+  }
+
+  async _ensureSkillsService() {
+    if (!this.skillsService) {
+      this.skillsService = getSkillsService(this.logger);
+      await this.skillsService.initialize();
+    }
+    return this.skillsService;
+  }
+
+  getDescription() {
+    return `Skills Tool: Browse and manage a global library of reusable skill instructions.
+
+Skills are structured knowledge packages containing instructions, checklists, templates, and reference files.
+Each skill is a directory with a skill.md file and optional supporting files.
+
+★ PROACTIVE USE — When a new task arrives or your focus shifts, your FIRST move should be:
+  \`{ "toolId": "skills", "action": "list" }\`
+Skim the names + descriptions. If anything looks relevant, "describe" it and follow its
+checklist instead of improvising. The team curates skills specifically so you don't have
+to re-derive recurring playbooks from scratch.
+
+PROGRESSIVE DISCLOSURE — Use this flow to minimize context usage:
+1. "list" → See all skills with section headings and sizes
+2. "describe" → Inspect a specific skill's structure in detail
+3. "read-section" → Load only the section you need
+4. "read" → Load full content only when necessary
+
+ACTIONS:
+
+1. LIST all skills:
+\`\`\`json
+{ "toolId": "skills", "action": "list" }
+\`\`\`
+Returns: name, description, section headings, line count, file count for each skill.
+
+2. DESCRIBE a skill (metadata only, no content):
+\`\`\`json
+{ "toolId": "skills", "action": "describe", "name": "code-review" }
+\`\`\`
+Returns: description, sections with line ranges, file list, size.
+
+3. READ full skill content:
+\`\`\`json
+{ "toolId": "skills", "action": "read", "name": "code-review" }
+\`\`\`
+Returns: full skill.md content + list of files in the skill directory.
+
+4. READ a specific SECTION:
+\`\`\`json
+{ "toolId": "skills", "action": "read-section", "name": "code-review", "section": "Checklist" }
+\`\`\`
+Returns: only the content under the specified ## heading.
+
+5. READ a supporting FILE:
+\`\`\`json
+{ "toolId": "skills", "action": "read-file", "name": "email-templates", "file": "templates/welcome.html" }
+\`\`\`
+Returns: content of a specific file within the skill directory.
+
+6. CREATE a new skill:
+\`\`\`json
+{ "toolId": "skills", "action": "create", "name": "my-skill", "content": "# My Skill\\n\\nInstructions here...\\n\\n## Section One\\n..." }
+\`\`\`
+Optional: "files" array of { "path": "relative/path.ext", "content": "..." } for supporting files.
+
+7. UPDATE an existing skill:
+\`\`\`json
+{ "toolId": "skills", "action": "update", "name": "my-skill", "content": "# Updated content..." }
+\`\`\`
+Optional: "files" array to add/update supporting files.
+
+8. DELETE a skill:
+\`\`\`json
+{ "toolId": "skills", "action": "delete", "name": "my-skill" }
+\`\`\`
+
+9. IMPORT a skill from disk:
+\`\`\`json
+{ "toolId": "skills", "action": "import", "source": "/path/to/skill-dir-or-file" }
+\`\`\`
+Optional: "name" to override the derived skill name. If source is a directory, it must contain a skill.md file.
+
+SKILL NAMING: Names must be kebab-case (lowercase, hyphens). Example: "code-review", "email-templates".`;
+  }
+
+  parseParameters(content) {
+    return content;
+  }
+
+  getRequiredParameters() {
+    return ['action'];
+  }
+
+  getSupportedActions() {
+    return Object.values(SKILLS_ACTIONS);
+  }
+
+  validateParameterTypes(params) {
+    const errors = [];
+    if (params.action && typeof params.action !== 'string') {
+      errors.push('action must be a string');
+    }
+    if (params.name !== undefined && typeof params.name !== 'string') {
+      errors.push('name must be a string');
+    }
+    if (params.content !== undefined && typeof params.content !== 'string') {
+      errors.push('content must be a string');
+    }
+    if (params.section !== undefined && typeof params.section !== 'string') {
+      errors.push('section must be a string');
+    }
+    if (params.file !== undefined && typeof params.file !== 'string') {
+      errors.push('file must be a string');
+    }
+    if (params.source !== undefined && typeof params.source !== 'string') {
+      errors.push('source must be a string');
+    }
+    return errors;
+  }
+
+  customValidateParameters(params) {
+    const errors = [];
+    const { action, name, content, section, file, source } = params;
+
+    const validActions = this.getSupportedActions();
+    if (!validActions.includes(action)) {
+      errors.push(`Invalid action: "${action}". Valid actions: ${validActions.join(', ')}`);
+      return errors;
+    }
+
+    // Action-specific required params
+    const needsName = [SKILLS_ACTIONS.DESCRIBE, SKILLS_ACTIONS.READ, SKILLS_ACTIONS.READ_SECTION, SKILLS_ACTIONS.READ_FILE, SKILLS_ACTIONS.CREATE, SKILLS_ACTIONS.UPDATE, SKILLS_ACTIONS.DELETE];
+    if (needsName.includes(action) && !name) {
+      errors.push(`"name" is required for action "${action}"`);
+    }
+    if (action === SKILLS_ACTIONS.CREATE && !content) {
+      errors.push('"content" is required for action "create"');
+    }
+    if (action === SKILLS_ACTIONS.READ_SECTION && !section) {
+      errors.push('"section" is required for action "read-section"');
+    }
+    if (action === SKILLS_ACTIONS.READ_FILE && !file) {
+      errors.push('"file" is required for action "read-file"');
+    }
+    if (action === SKILLS_ACTIONS.IMPORT && !source) {
+      errors.push('"source" is required for action "import"');
+    }
+
+    return errors;
+  }
+
+  async execute(params, context = {}) {
+    const service = await this._ensureSkillsService();
+    const { action } = params;
+
+    try {
+      switch (action) {
+        case SKILLS_ACTIONS.LIST:
+          return this._formatResult(await service.listSkills(), 'Skills listed');
+
+        case SKILLS_ACTIONS.DESCRIBE:
+          return this._formatResult(await service.describeSkill(params.name), `Skill described: ${params.name}`);
+
+        case SKILLS_ACTIONS.READ:
+          return this._formatResult(await service.readSkill(params.name), `Skill read: ${params.name}`);
+
+        case SKILLS_ACTIONS.READ_SECTION:
+          return this._formatResult(await service.readSkillSection(params.name, params.section), `Section read: ${params.section}`);
+
+        case SKILLS_ACTIONS.READ_FILE:
+          return this._formatResult(await service.readSkillFile(params.name, params.file), `File read: ${params.file}`);
+
+        case SKILLS_ACTIONS.CREATE:
+          return this._formatResult(await service.createSkill(params.name, params.content, params.files || [], params.description || null), `Skill created: ${params.name}`);
+
+        case SKILLS_ACTIONS.UPDATE:
+          return this._formatResult(await service.updateSkill(params.name, params.content || null, params.files || [], params.description || null), `Skill updated: ${params.name}`);
+
+        case SKILLS_ACTIONS.DELETE:
+          await service.deleteSkill(params.name);
+          return this._formatResult({ deleted: params.name }, `Skill deleted: ${params.name}`);
+
+        case SKILLS_ACTIONS.IMPORT:
+          return this._formatResult(await service.importSkill(params.source, params.name || null, params.description || null), `Skill imported: ${params.name || params.source}`);
+
+        default:
+          return { success: false, error: `Unknown action: ${action}` };
+      }
+    } catch (error) {
+      return { success: false, error: error.message };
+    }
+  }
+
+  _formatResult(data, message) {
+    return {
+      success: true,
+      result: data,
+      message
+    };
+  }
+
+  getParameterSchema() {
+    return {
+      type: 'object',
+      required: ['action'],
+      properties: {
+        action: {
+          type: 'string',
+          enum: Object.values(SKILLS_ACTIONS),
+          description: 'The skill action to perform'
+        },
+        name: {
+          type: 'string',
+          description: 'Skill name (kebab-case)'
+        },
+        content: {
+          type: 'string',
+          description: 'Skill content (markdown)'
+        },
+        section: {
+          type: 'string',
+          description: 'Section heading to read (for read-section action)'
+        },
+        file: {
+          type: 'string',
+          description: 'Relative file path within skill directory (for read-file action)'
+        },
+        source: {
+          type: 'string',
+          description: 'Source file or directory path (for import action)'
+        },
+        files: {
+          type: 'array',
+          items: {
+            type: 'object',
+            properties: {
+              path: { type: 'string' },
+              content: { type: 'string' }
+            }
+          },
+          description: 'Additional supporting files (for create/update actions)'
+        }
+      }
+    };
+  }
+}
+
+export default SkillsTool;

package/src/utilities/constants.js

@@ -26,8 +26,26 @@ const SYSTEM_DEFAULTS = {
 };
 
 // Model Router Configuration
+//
+// ROUTER_MODEL is the model the Dynamic Model Routing feature calls
+// (via a cheap chat-completion request) to decide which "real" model
+// should handle each turn. Resolution order:
+//   1. env LOXIA_ROUTER_MODEL — operator override, no rebuild needed
+//   2. 'gpt-4.1-nano' — current live default. The platform's
+//      autopilot-model-router deployment uses gpt-4.1-nano as its
+//      underlying model, and the model-catalog keys entries by the
+//      underlying model name (NOT the Azure deployment name), so the
+//      CLI must ask for 'gpt-4.1-nano' to be matched. Cheaper than the
+//      retired OpenAI 'model-router' product, same job.
+//
+// Historical note: this used to be the literal string 'model-router',
+// matching an OpenAI product name. That product is no longer in our
+// Azure catalog (no deployment keyed under that name), which caused
+// every routing call to fail with HTTP 400 "Unsupported model:
+// model-router" until the circuit breaker tripped. The fix migrates
+// the default to the underlying model name that IS in the catalog.
 const MODEL_ROUTER_CONFIG = {
-  ROUTER_MODEL: 'model-router', // Autopilot model router deployment
+  ROUTER_MODEL: process.env.LOXIA_ROUTER_MODEL || 'gpt-4.1-nano',
   CONTEXT_MESSAGES_COUNT: 5, // Number of recent messages to include
   BENCHMARK_REFRESH_INTERVAL: 3600000, // 1 hour in milliseconds
   FALLBACK_ON_ERROR: true, // Continue with previous model on router error