fix: decouple critical_actions from hasSidecar configuration

critical_actions is for ANY activation behavior (loading files, startup
tasks, data fetches), not just for agents with sidecars.

Key changes:
- Template: Changed `has_sidecar` to `has_critical_actions` condition
- Validation: Agents WITHOUT sidecar CAN have critical_actions
- Validation: Agents WITH sidecar MUST have critical_actions (for loading)
- critical-actions.md: Clarified both configurations with examples
- agent-compilation.md: Updated with both patterns
- Edit flow: When removing sidecar, conditionally keep critical_actions
- understanding-agent-types.md: Added comparison table row for critical_actions

This corrects the misconception that only agents with sidecars have
critical_actions - in reality, any agent may have activation behaviors.

Author: Brian Madison

src/workflows/agent/data/agent-architecture.md

@@ -323,8 +323,8 @@ Two forms:
 
 - [ ] Under ~250 lines
 - [ ] No external dependencies
-- [ ] No `critical_actions` section
 - [ ] No sidecar path references
+- [ ] If critical_actions present, no sidecar file references
 
 ### WITH Sidecar (hasSidecar: true)
 
@@ -345,7 +345,7 @@ Two forms:
 4. **Sensible defaults** in install_config (if using)
 5. **Numbered steps** in multi-step prompts
 6. **Keep under ~250 lines** for agents without sidecar
-7. **critical_actions MANDATORY** for agents with sidecar
+7. **critical_actions** for any activation behavior (sidecar or other)
 8. **Enforce domain restrictions** for agents with sidecar
 9. **Reference past naturally** - Don't dump memory
 10. **Design for growth** - Structure for accumulation (with sidecar)

src/workflows/agent/data/agent-compilation.md

@@ -40,10 +40,10 @@ agent:
     principles:
       - "Core belief or methodology"
 
-  critical_actions:              # Optional - for Expert agents only
-    - "Load ./sidecar/memories.md"
-    - "Load ./sidecar/instructions.md"
-    - "ONLY access ./sidecar/"
+  critical_actions:              # Optional - for any agent with activation behavior
+    - "Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/memories.md"
+    - "Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/instructions.md"
+    - "ONLY read/write files in {project-root}/_bmad/_memory/journal-sidecar/"
 
   prompts:                        # Optional - for Simple/Expert agents
     - id: prompt-name
@@ -220,34 +220,52 @@ When building agent YAML, **DO NOT:**
 **DO:**
 - [ ] Define metadata (id, name, title, icon, module)
 - [ ] Define persona (role, identity, communication_style, principles)
-- [ ] Define critical_actions (Expert agents only)
+- [ ] Define critical_actions (if agent has activation behavior)
 - [ ] Define prompts with IDs (Simple/Expert agents only)
 - [ ] Define menu with your custom items only
 - [ ] Use proper trigger format: `XX or fuzzy match on command-name`
 - [ ] Use proper description format: `[XX] Description text`
 
 ---
 
-## Expert Agent: critical_actions
+## Agent critical_actions
 
-For Expert agents with sidecars, your `critical_actions` become activation steps:
+For any agent with activation behavior, your `critical_actions` become activation steps.
+
+### Agents with sidecar (hasSidecar: true):
 
 ```yaml
 critical_actions:
-  - "Load COMPLETE file ./agent-sidecar/memories.md"
-  - "Load COMPLETE file ./agent-sidecar/instructions.md"
-  - "ONLY read/write files in ./agent-sidecar/"
+  - "Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/memories.md"
+  - "Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/instructions.md"
+  - "ONLY read/write files in {project-root}/_bmad/_memory/journal-sidecar/"
 ```
 
 The compiler injects these as steps 4, 5, 6 in the activation block:
 
 ```xml
-<step n="4">Load COMPLETE file ./agent-sidecar/memories.md</step>
-<step n="5">Load COMPLETE file ./agent-sidecar/instructions.md</step>
-<step n="6">ONLY read/write files in ./agent-sidecar/</step>
+<step n="4">Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/memories.md</step>
+<step n="5">Load COMPLETE file {project-root}/_bmad/_memory/journal-sidecar/instructions.md</step>
+<step n="6">ONLY read/write files in {project-root}/_bmad/_memory/journal-sidecar/</step>
 <step n="7">ALWAYS communicate in {communication_language}</step>
 ```
 
+### Agents without sidecar (hasSidecar: false):
+
+```yaml
+critical_actions:
+  - "Give user an inspirational quote before showing menu"
+  - "Review {project-root}/finances/ for most recent data file"
+```
+
+The compiler injects these as steps 4, 5 in the activation block:
+
+```xml
+<step n="4">Give user an inspirational quote before showing menu</step>
+<step n="5">Review {project-root}/finances/ for most recent data file</step>
+<step n="6">ALWAYS communicate in {communication_language}</step>
+```
+
 ---
 
 ## Division of Responsibilities
@@ -269,5 +287,5 @@ The compiler injects these as steps 4, 5, 6 in the activation block:
 - **Focus on:** Clean YAML structure, persona definition, custom menu items
 - **Ignore:** What happens after compilation—that's the compiler's job
 - **Remember:** Every agent gets MH, CH, PM, DA automatically—don't add them
-- **Expert agents:** Use `critical_actions` for sidecar file loading
+- **critical_actions:** Use for activation behavior (loading files, greetings, data fetches)
 - **Module agents:** Use `workflow:` or `exec:` references, not inline actions

src/workflows/agent/data/agent-validation.md

@@ -82,14 +82,20 @@ Validate agents meet BMAD quality standards. The validation approach depends on
 
 - [ ] `agent.metadata.hasSidecar` is `false`
 - [ ] Single .agent.yaml file (no sidecar folder)
-- [ ] All content contained in YAML (no external file dependencies)
-- [ ] No `critical_actions` section
+- [ ] No `{project-root}/_bmad/_memory/` paths to sidecar folders
 - [ ] Total size under ~250 lines (unless justified)
 
+### critical_actions Validation (OPTIONAL)
+
+- [ ] If present, `critical_actions` contains activation behaviors
+- [ ] No references to sidecar files (no sidecar exists)
+- [ ] No placeholder text in critical_actions
+- [ ] No compiler-injected steps (Load persona, Load config, greeting, etc.)
+
 ### Path Validation
 
 - [ ] No sidecar paths present
-- [ ] No `{project-root}/_bmad/_memory/` paths
+- [ ] No `{project-root}/_bmad/_memory/` paths to sidecar folders
 
 ### Reference Comparison
 
@@ -102,7 +108,7 @@ Validate agents meet BMAD quality standards. The validation approach depends on
 ### Structure Validation
 
 - [ ] `agent.metadata.hasSidecar` is `true`
-- [ ] `agent.critical_actions` exists (MANDATORY)
+- [ ] `agent.metadata.sidecar-folder` is specified
 - [ ] Sidecar folder exists: `{agent-name}-sidecar/`
 - [ ] Folder name matches agent name
 - [ ] `instructions.md` exists in sidecar (recommended)
@@ -113,7 +119,7 @@ Validate agents meet BMAD quality standards. The validation approach depends on
 - [ ] **communication_style** includes memory reference patterns
 - [ ] Memory references feel natural: "Last time you mentioned..." or "I've noticed patterns..."
 
-### critical_actions Validation (MANDATORY)
+### critical_actions Validation (MANDATORY for hasSidecar: true)
 
 - [ ] `critical_actions` section exists
 - [ ] Contains at minimum 3 actions
@@ -206,6 +212,15 @@ critical_actions:
   - 'ONLY read/write files in {project-root}/_bmad/_memory/{sidecar-folder}/'
 ```
 
+### Issue: critical_actions for hasSidecar: false
+
+**Note:** Agents without sidecar can have critical_actions for other activation behaviors:
+```yaml
+critical_actions:
+  - 'Give user an inspirational quote before showing menu'
+  - 'Review {project-root}/finances/ for most recent data file'
+```
+
 ### Issue: Communication Style Missing Memory References (hasSidecar: true)
 
 **Fix:** Add memory reference patterns: "I reference past naturally: 'Last time you mentioned...'"

src/workflows/agent/data/critical-actions.md

@@ -14,14 +14,20 @@ Numbered steps that execute FIRST when an agent activates.
 - Startup behavior (greeting enhancement, data fetch, state init)
 - Any MUST-do activation behavior
 
-**Applies to:** BOTH Simple and Expert agents
+**Applies to:** BOTH agents with and without sidecar
 
 ---
 
-## Expert Agent Pattern
+## Agent WITH Sidecar Pattern (hasSidecar: true)
+
+Agents with sidecars use `critical_actions` to:
+1. Load sidecar memory files (memories.md, instructions.md)
+2. Set file access boundaries (restrict to sidecar folder only)
+3. Optionally write to memory during activation if needed
+4. Add any other startup behaviors specific to the agent
 
 ```yaml
-# ✅ CORRECT Expert Agent
+# ✅ CORRECT Agent with sidecar
 critical_actions:
   - 'Load COMPLETE file {project-root}/_bmad/_memory/journal-keeper-sidecar/memories.md'
   - 'Load COMPLETE file {project-root}/_bmad/_memory/journal-keeper-sidecar/instructions.md'
@@ -34,15 +40,20 @@ critical_actions:
 - Sidecar created next to agent.yaml during BUILD, then copied to `_memory/` during BMAD INSTALLATION
 - Use `{project-root}/_bmad/_memory/{sidecar-folder}/` format for RUNTIME paths in agent YAML
 
+**MANDATORY for hasSidecar: true:** At minimum 3 actions for loading memories, instructions, and restricting file access.
+
 ---
 
-## Simple Agent Pattern
+## Agent WITHOUT Sidecar Pattern (hasSidecar: false)
+
+Agents without sidecars may use `critical_actions` for other activation behaviors:
 
 ```yaml
-# ✅ CORRECT Simple Agent with activation behavior
+# ✅ CORRECT Agent without sidecar, with activation behavior
 critical_actions:
   - 'Give user an inspirational quote before showing menu'
   - 'Review {project-root}/finances/ for most recent data file'
+  - 'Fetch latest stock prices before displaying menu'
 ```
 
 **Note:** Agents without activation needs can omit `critical_actions` entirely.
@@ -53,8 +64,8 @@ critical_actions:
 
 | Type | Pattern |
 |------|---------|
-| Expert sidecar | `{project-root}/_bmad/_memory/{sidecar-folder}/file.md` |
-| Simple data | `{project-root}/finances/data.csv` |
+| Sidecar memory | `{project-root}/_bmad/_memory/{sidecar-folder}/file.md` |
+| Other data | `{project-root}/finances/data.csv` |
 | Output folders | `{output_folder}/results/` |
 
 ---

src/workflows/agent/data/understanding-agent-types.md

@@ -70,6 +70,10 @@ agent-name.agent.yaml (~250 lines max)
 - Motivational Gym Bro (hype up your workout)
 - Sassy Fortune Teller (mystical snark)
 
+**Optional critical_actions:**
+- Can have critical_actions for activation behaviors (quotes, data fetches, etc.)
+- Must NOT reference sidecar files (no sidecar exists)
+
 **Reference:** `./data/reference/simple-examples/commit-poet.agent.yaml`
 
 ---
@@ -124,6 +128,7 @@ critical_actions:
 | ----------------- | ------------------------ | ------------------------------ |
 | File structure    | Single YAML (~250 lines) | YAML + sidecar/ (files)       |
 | Persistent memory | No                       | Yes                            |
+| critical_actions  | Optional (other behaviors) | MANDATORY (load sidecar files) |
 | Custom workflows  | Inline prompts           | Sidecar workflows (on-demand)  |
 | File access       | Project/output           | Restricted to sidecar domain  |
 | Session state     | Stateless                | Remembers across sessions      |

src/workflows/agent/steps-c/step-06-activation.md

@@ -165,12 +165,21 @@ Ask user:
 - Each action should be clear and scoped
 - Document rationale for each
 
-**For agents WITH sidecar, critical_actions typically include:**
+**For agents WITH sidecar, critical_actions MUST include:**
 ```
 - "Load COMPLETE file {project-root}/_bmad/_memory/{sidecar-folder}/memories.md"
 - "Load COMPLETE file {project-root}/_bmad/_memory/{sidecar-folder}/instructions.md"
 - "ONLY read/write files in {project-root}/_bmad/_memory/{sidecar-folder}/ - private space"
 ```
+Plus any additional activation behaviors the agent needs.
+
+**For agents WITHOUT sidecar, critical_actions are OPTIONAL and can include:**
+```
+- "Give user an inspirational quote before showing menu"
+- "Fetch latest data from {project-root}/finances/ before displaying menu"
+- "Display a quick status summary on activation"
+```
+Agents without sidecar omit critical_actions entirely if no activation behavior is needed.
 
 **If omitting:**
 - State clearly: "This agent will not have critical_actions"

src/workflows/agent/steps-e/e-04-sidecar-metadata.md

@@ -57,8 +57,8 @@ If user wants metadata changes:
 **For sidecar conversion:**
 - "Converting from hasSidecar: {current} to {target}"
 - Explain implications:
-  - false → true: Need to create sidecar folder, add critical_actions
-  - true → false: Remove sidecar fields, remove critical_actions
+  - false → true: Need to create sidecar folder, add critical_actions with sidecar file loading
+  - true → false: Remove sidecar fields; if critical_actions only has sidecar references, remove section; otherwise keep non-sidecar critical_actions
 - Update editPlan with conversion
 
 **For metadata field changes:**

src/workflows/agent/steps-e/e-08-edit-agent.md

@@ -104,7 +104,8 @@ For each planned edit:
 **true → false (Removing sidecar):**
 - Set `hasSidecar: false`
 - Remove `metadata.sidecar-folder` and `metadata.sidecar-path`
-- Remove `critical_actions` section
+- If critical_actions contains only sidecar references, remove the section
+- If critical_actions contains non-sidecar activation behaviors, keep and clean sidecar references
 - Remove sidecar references from menu actions
 - Optionally archive sidecar folder
 

src/workflows/agent/steps-v/v-02d-validate-structure.md

@@ -79,7 +79,7 @@ Perform these checks systematically - validate EVERY rule specified in agentComp
 **For Agents WITHOUT Sidecar (hasSidecar is false):**
 - [ ] No sidecar requirements
 - [ ] No sidecar-folder path in metadata
-- [ ] No critical_actions section
+- [ ] If critical_actions present, no sidecar file references
 - [ ] Menu handlers use only internal references (#) or inline prompts
 - [ ] Total size under ~250 lines (unless justified)
 

src/workflows/agent/steps-v/v-02e-validate-sidecar.md

@@ -89,7 +89,7 @@ For each sidecar path reference in agent YAML:
 **IF hasSidecar = false:**
 - [ ] Mark sidecar validation as N/A
 - [ ] Confirm no sidecar-folder path in metadata
-- [ ] Confirm no critical_actions section
+- [ ] Confirm no sidecar references in critical_actions (if present)
 - [ ] Confirm no sidecar references in menu handlers
 
 ### 3. Append Findings to Report