Update declarative agent skill for schema v1.8

.claude-plugin/marketplace.json

@@ -20,7 +20,7 @@
     {
       "name": "microsoft-365-agents-toolkit",
       "description": "Toolkit for building and evaluating Microsoft 365 Copilot declarative agents \u2014 scaffolding, JSON manifest development, capability configuration, and eval workflows.",
-      "version": "1.3.1",
+      "version": "1.5.0",
       "source": "./plugins/microsoft-365-agents-toolkit"
     },
     {

marketplace.json

@@ -21,7 +21,7 @@
       "name": "microsoft-365-agents-toolkit",
       "source": "./plugins/microsoft-365-agents-toolkit",
       "description": "Toolkit for building and evaluating Microsoft 365 Copilot declarative agents \u2014 scaffolding, JSON manifest development, capability configuration, and eval workflows.",
-      "version": "1.3.1"
+      "version": "1.5.0"
     },
     {
       "name": "workiq-productivity",

plugins/microsoft-365-agents-toolkit/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "microsoft-365-agents-toolkit",
-  "version": "1.3.1",
+  "version": "1.5.0",
   "description": "Toolkit for building and evaluating Microsoft 365 Copilot declarative agents \u2014 scaffolding, JSON manifest development, capability configuration, and eval workflows.",
   "skills": "./skills/"
 }

plugins/microsoft-365-agents-toolkit/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "microsoft-365-agents-toolkit",
-  "version": "1.3.1",
+  "version": "1.5.0",
   "description": "Toolkit for building and evaluating Microsoft 365 Copilot declarative agents \u2014 scaffolding, JSON manifest development, capability configuration, and eval workflows.",
   "skills": "./skills/"
 }

plugins/microsoft-365-agents-toolkit/.github/plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "microsoft-365-agents-toolkit",
-  "version": "1.3.1",
+  "version": "1.5.0",
   "description": "Toolkit for building and evaluating Microsoft 365 Copilot declarative agents \u2014 scaffolding, JSON manifest development, capability configuration, and eval workflows.",
   "skills": "./skills/"
 }

plugins/microsoft-365-agents-toolkit/skills/declarative-agent-developer/SKILL.md

@@ -10,6 +10,8 @@ description: >
   "configure my agent", "deploy my agent", "fix my agent manifest", "edit my agent",
   "localize my agent", "add localization", "translate my agent", "multi-language agent",
   "add an API plugin", "add an MCP plugin", "add OAuth to my plugin",
+  "add email actions to agent", "add meeting actions to agent",
+  "add editorial answers to agent", "set agent response mode",
   "review instructions", "improve instructions", "fix my instructions"
 ---
 
@@ -67,7 +69,7 @@ When you encounter ANY problem (missing files, malformed JSON, validation errors
 - Missing `declarativeAgent.json` → Detect (file not found) → Inform ("no manifest found") → Ask ("would you like to create a new agent?")
 - Malformed JSON → Detect (parse errors) → Inform (list specific syntax issues) → Ask ("should I fix these syntax errors?")
 - Validation errors → Detect (parse and check manifest) → Inform (list all errors) → Ask ("how would you like to fix these?")
-- Version incompatibility → Detect (feature requires newer version) → Inform ("this feature requires v1.6, your agent is v1.4") → Ask ("should I upgrade?")
+- Version incompatibility → Detect (feature requires newer version) → Inform ("this feature requires v1.8, your agent is v1.7") → Ask ("should I upgrade?")
 
 ---
 
@@ -80,6 +82,10 @@ When you encounter ANY problem (missing files, malformed JSON, validation errors
 | Adding an API plugin | [API Plugins](references/api-plugins.md) |
 | Adding an MCP server | [MCP Plugin](references/mcp-plugin.md) |
 | Adding OAuth to an MCP or API plugin | [Authentication](references/authentication.md) |
+| Adding email write operations (`EmailActions`) | [Editing Workflow](references/editing-workflow.md) |
+| Adding meeting or calendar write operations (`MeetingActions`) | [Editing Workflow](references/editing-workflow.md) |
+| Adding predefined semantic Q&A (`editorial_answers`) | [Editing Workflow](references/editing-workflow.md) |
+| Choosing the default response mode | [Conversation Design](references/conversation-design.md) |
 | Reviewing or improving existing agent instructions | [Instruction Review](references/instruction-review.md) |
 | User reports agent gives generic/wrong answers | [Instruction Review](references/instruction-review.md) |
 | Localizing an agent into multiple languages | [Localization](references/localization.md) |
@@ -88,6 +94,21 @@ When you encounter ANY problem (missing files, malformed JSON, validation errors
 
 ---
 
+## Build-Time Feature Discovery
+
+When creating or designing an agent, actively evaluate these features instead of waiting for the user to name their manifest identifiers:
+
+| Requirement signal | Feature to surface |
+|--------------------|--------------------|
+| Send, archive, flag, move, delete, or organize email; manage rules, auto-replies, or folders | `EmailActions` |
+| Schedule meetings, create calendar events or time-finding polls, or provide time insights | `MeetingActions` |
+| Return approved, stable answers for semantically similar FAQ-style questions | `editorial_answers` |
+| Prefer fast replies or deeper reasoning by default | `behavior_overrides.default_response_mode` |
+
+Surface only relevant options during requirements discovery. Do not add them without clear user intent. For write capabilities, distinguish them from the read-only `Email` and `Meetings` capabilities and include confirmation guidance for sensitive operations.
+
+---
+
 ## ATK CLI Setup
 
 Before running any ATK commands, check if the ATK CLI is available by running `npx -y --package @microsoft/m365agentstoolkit-cli atk --version`. If not found, **STOP and tell the user** — do NOT attempt to install it yourself.
@@ -133,7 +154,9 @@ Then read `M365_TITLE_ID` from `env/.env.local` and **ALWAYS** present the revie
 Before adding ANY feature, read the `version` field in `declarativeAgent.json` and check the [Schema](references/schema.md) feature matrix. If the feature isn't supported in that version, **refuse** and offer to upgrade.
 
 Key version gates:
-- `sensitivity_label`, `worker_agents`, `EmbeddedKnowledge` → **v1.6 only**
+- `EmailActions`, `MeetingActions` → **v1.8+**
+- `editorial_answers`, `behavior_overrides.default_response_mode`, `conversation_starters[].depends_on` → **v1.7+**
+- `sensitivity_label`, `worker_agents`, `EmbeddedKnowledge`, `user_overrides` → **v1.6+**
 - `Meetings` → **v1.5+**
 - `ScenarioModels`, `behavior_overrides`, `disclaimer` → **v1.4+**
 - `Dataverse`, `TeamsMessages`, `Email`, `People` → **v1.3+**

plugins/microsoft-365-agents-toolkit/skills/declarative-agent-developer/references/api-plugins.md

@@ -693,7 +693,7 @@ Reference the API plugin in your `declarativeAgent.json`:
 
 ```json
 {
-  "version": "v1.6",
+  "version": "v1.8",
   "name": "Repairs Agent",
   "description": "An agent to manage repair tickets",
   "instructions": "You help users manage repair tickets. Use the repairs API to search, create, update, and delete repairs.",

plugins/microsoft-365-agents-toolkit/skills/declarative-agent-developer/references/best-practices.md

@@ -10,6 +10,7 @@ Follow these best practices for successful M365 Copilot agent development.
 - **PII Handling:** Follow data protection regulations when handling personal information
 - **Audit Logging:** Implement comprehensive audit trails for all agent actions
 - **Secret Storage:** Never hardcode credentials; use Azure Key Vault or environment variables
+- **Write Capability Confirmation:** Require clear user confirmation before sensitive `EmailActions` or `MeetingActions`, such as sending or deleting email and creating or changing calendar events
 
 ## Performance
 
@@ -18,6 +19,7 @@ Follow these best practices for successful M365 Copilot agent development.
 - **Caching Strategy:** Implement appropriate caching for frequently accessed data
 - **Response Time:** Keep operations under 30 seconds to avoid timeouts
 - **Batch Operations:** Use batch APIs when processing multiple items
+- **Response Mode Selection:** Use `Quick response` for latency-sensitive agents, `Think deeper` for complex analytical agents, and `Auto` for mixed workloads
 
 ## Error Handling
 
@@ -30,6 +32,8 @@ Follow these best practices for successful M365 Copilot agent development.
 ## Testing
 
 - **Test All Conversation Starters:** Verify each starter works as intended
+- **Test Write Actions:** Verify confirmation, permissions, cancellation, and failure behavior for `EmailActions` and `MeetingActions`
+- **Test Editorial Matching:** Check close paraphrases and unrelated questions against every `editorial_answers` entry
 - **Test Edge Cases:** Test with missing data, invalid inputs, and error conditions
 - **Security Testing:** Verify scoping and permission controls
 - **Cross-Environment Testing:** Test in dev, staging, and production environments
@@ -50,6 +54,7 @@ Follow these best practices for successful M365 Copilot agent development.
 - **Code Organization:** Separate concerns (capabilities, API plugins, models)
 - **Version Control:** Use semantic versioning for shared agents
 - **Change Management:** Document changes and maintain changelog
+- **Editorial Answer Governance:** Assign an owner and review cadence for approved answers so stale canonical content is removed promptly
 
 ## Conversation Design
 

plugins/microsoft-365-agents-toolkit/skills/declarative-agent-developer/references/conversation-design.md

@@ -108,12 +108,14 @@ Describe the purpose and context for each data source the agent can use. You don
 - **Copilot connector knowledge**: Reference by connector name — "Use `ServiceNow KB` for help articles."
 - **SharePoint/OneDrive**: Describe the data source — "Search internal HR policy documents" or "Reference company documents in SharePoint."
 - **Email**: Describe the scenario — "Check user emails for relevant information."
+- **Email actions**: Describe when to send, archive, flag, move, delete, or organize email, and require confirmation before sensitive writes.
 - **Teams messages**: "Search Teams channels and chat messages." (messages only — NOT transcripts)
 - **Meetings**: "Check calendar events, attendees, and meeting transcripts." (transcripts come from `Meetings`, not `TeamsMessages`)
+- **Meeting actions**: Describe when to schedule or change calendar events, create time-finding polls, or surface time insights, and require confirmation before calendar writes.
 - **Code interpreter**: "Use code interpreter to generate charts."
 - **People knowledge**: "Look up people in the organization for contact info."
 
-> **Common mistake:** Assuming meeting transcripts are part of `TeamsMessages`. Transcripts are retrieved through the `Meetings` capability. `TeamsMessages` covers channel posts, DMs, and meeting chat messages only. See the [Capability Reference](instruction-review.md#capability-reference-v16) for the full mapping.
+> **Common mistake:** Assuming meeting transcripts are part of `TeamsMessages`. Transcripts are retrieved through the `Meetings` capability. `TeamsMessages` covers channel posts, DMs, and meeting chat messages only. See the [Capability Reference](instruction-review.md#capability-reference-v18) for the full mapping.
 
 ### 8. Provide Examples
 
@@ -163,6 +165,54 @@ Developing instructions is an iterative process:
 
 ---
 
+## Manifest-Level Experience Controls
+
+Evaluate these options while designing a new agent; they are easy to miss if discovery focuses only on knowledge sources and plugins.
+
+### Editorial answers (v1.7+)
+
+Use `editorial_answers` for approved, stable answers that should be selected through semantic similarity, such as official support contacts or policy FAQs.
+
+- Use it for canonical content that changes infrequently.
+- Do not use it for personalized, permission-sensitive, or rapidly changing answers.
+- Keep each question and answer focused on one intent.
+- Review inline answers as governed content because they reduce normal generative variation.
+
+```json
+{
+  "editorial_answers": {
+    "answers": [
+      {
+        "question": "How do I contact the help desk?",
+        "answer": "Email helpdesk@contoso.com or call extension 5555."
+      }
+    ]
+  }
+}
+```
+
+### Default response mode (v1.7+)
+
+Choose `behavior_overrides.default_response_mode` from the dominant workload:
+
+| Mode | Use when |
+|------|----------|
+| `Auto` | Requests vary between simple lookups and complex analysis. This is the default. |
+| `Quick response` | Low latency matters more than deep reasoning, such as routing or short FAQ agents. |
+| `Think deeper` | The agent primarily performs complex analysis, planning, or multi-step decisions. |
+
+Users can override the default in the model selector. Do not rely on the setting as a hard enforcement mechanism.
+
+```json
+{
+  "behavior_overrides": {
+    "default_response_mode": "Think deeper"
+  }
+}
+```
+
+---
+
 ## Example Instructions
 
 The following example is for an IT help desk agent:
@@ -369,14 +419,29 @@ Identify the top 3-5 tasks users will perform:
 }
 ```
 
-#### 4. Use Natural Language
+#### 4. Gate Capability-Specific Starters (v1.7+)
+
+Use `depends_on` when a starter only works with specific configured capabilities. Every dependency must be present before the starter is shown.
+
+```json
+{
+  "title": "Project Update",
+  "text": "Summarize internal email and public web updates for this project.",
+  "depends_on": [
+    { "name": "capabilities", "id": "Email" },
+    { "name": "capabilities", "id": "WebSearch" }
+  ]
+}
+```
+
+#### 5. Use Natural Language
 
 Write starters as users would naturally speak:
 
 ✅ Good: "What's the latest on Project Phoenix?"
 ❌ Bad: "Query project status for Phoenix"
 
-#### 5. Provide 3-6 Starters (Not Too Many)
+#### 6. Provide 3-6 Starters (Not Too Many)
 - **Too few** (<3): Users don't see the full capability range
 - **Just right** (3-6): Good variety without overwhelming
 - **Too many** (>6): Clutters UI, users won't read them all

plugins/microsoft-365-agents-toolkit/skills/declarative-agent-developer/references/editing-workflow.md

@@ -65,6 +65,9 @@ When you add, remove, or modify ANY capability or plugin, you MUST complete ALL
 - Identify the agent's primary purpose and target users
 - Determine required data sources (M365 services, external APIs)
 - List necessary actions the agent must perform
+- Identify email and calendar **write** operations separately from read/search requirements
+- Identify approved, stable FAQ answers that should use canonical response content
+- Decide whether the default experience should favor automatic selection, low latency, or deeper reasoning
 - Identify security and compliance requirements
 
 ### Step 2: Design the Agent Architecture
@@ -76,6 +79,15 @@ When you add, remove, or modify ANY capability or plugin, you MUST complete ALL
 - Plan authentication and authorization strategy
 - Design conversation flow and instructions
 
+**Build-time feature discovery:**
+
+| User requirement | Manifest feature | Design rule |
+|------------------|------------------|-------------|
+| Email triage, supervised send, delete, rules, auto-reply, or folder management | `EmailActions` (v1.8+) | This is independent of read-only `Email`; its scopes do not carry over. Add confirmation guidance for sensitive writes. |
+| Meeting scheduling, calendar events, time-finding polls, or time insights | `MeetingActions` (v1.8+) | This is separate from read-only `Meetings`. Add confirmation guidance before creating or changing events. |
+| Canonical answers for semantically similar questions | `editorial_answers` (v1.7+) | Use for approved, stable Q&A. Do not use for dynamic or personalized data. |
+| A consistent default speed/reasoning experience | `behavior_overrides.default_response_mode` (v1.7+) | Choose `Auto`, `Quick response`, or `Think deeper` based on the dominant workload. |
+
 ### Step 3: Edit JSON Manifest Files
 
 **⚠️ PRE-EDIT CHECK — Before making ANY edits, do ALL of these:**
@@ -87,7 +99,7 @@ When you add, remove, or modify ANY capability or plugin, you MUST complete ALL
    - Only after the user confirms, fix with surgical edits, then re-read the file
    - Then continue with the user's original request as a separate step
 
-2. **Check the schema version**: Read the `"version"` field in `declarativeAgent.json` (e.g., `"v1.4"`, `"v1.6"`). For EVERY feature you plan to add, verify it exists in that version using the [feature matrix](schema.md). If a requested feature requires a newer version → **STOP. Tell the user.** Offer to upgrade the version first.
+2. **Check the schema version**: Read the `"version"` field in `declarativeAgent.json` (e.g., `"v1.4"`, `"v1.8"`). For EVERY feature you plan to add, verify it exists in that version using the [feature matrix](schema.md). If a requested feature requires a newer version → **STOP. Tell the user.** Offer to upgrade the version first.
 
 3. **Run a proactive instruction review** (if the edit touches instructions or capabilities): Before modifying instructions or adding/removing capabilities, run [Instruction Review](instruction-review.md) **Phase 1 (Inventory)**, **Phase 2 (Comprehension Check)**, and **Phase 3 (Diagnose)** against the current instructions. This catches existing problems before you add to them. For Phase 2, use the brief confirmation shortcut ("I see this agent is designed to [purpose]…") since this is a proactive check. If the review finds high-severity issues (C1, C3, C11, D1-D8), inform the user and offer to fix them as part of the current edit.