feat: add agent definitions API for managing Codex and Claude agent configurations

Author: catlog22

.codex/agents/cli-execution-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are an intelligent CLI execution specialist that autonomously orchestrates context discovery and optimal tool execution.
 
@@ -331,4 +331,4 @@ Codex unavailable → Gemini/Qwen write mode
 - `claude-module-unified.txt` - Universal module/file documentation
 
 ---
-"""
+'''

.codex/agents/cli-explore-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "read-only"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are a specialized CLI exploration agent that autonomously analyzes codebases and generates structured outputs.
 
@@ -229,4 +229,4 @@ Brief summary:
 **Consumption Pattern**:
 - Plan phase: Fully consumes `exploration-notes.md`
 - Execute phase: Consumes `exploration-notes-refined.md`, reduced noise, improved efficiency
-"""
+'''

.codex/agents/cli-lite-planning-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are a generic planning agent that generates structured plan JSON for lite workflows. Output format is determined by the schema reference provided in the prompt. You execute CLI planning tools (Gemini/Qwen), parse results, and generate planObject conforming to the specified schema.
 
@@ -898,4 +898,4 @@ After Phase 4 planObject generation:
 5. **Return** → Plan with `_metadata.quality_check` containing execution result
 
 **CLI Fallback**: Gemini → Qwen → Skip with warning (if both fail)
-"""
+'''

.codex/agents/cli-planning-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are a specialized execution agent that bridges CLI analysis tools with task generation. You execute Gemini/Qwen CLI commands for failure diagnosis, parse structured results, and dynamically generate task JSON files for downstream execution.
 
@@ -550,4 +550,4 @@ See: `.process/iteration-{iteration}-cli-output.txt`
      estimated_complexity: "medium"
    }
    ```
-"""
+'''

.codex/agents/code-developer.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are a code execution specialist focused on implementing high-quality, production-ready code. You receive tasks with context and execute them efficiently using strict development standards.
 
@@ -505,4 +505,4 @@ Before completing any task, verify:
 - Document all new interfaces, types, and constants for dependent task reference
 ### Windows Path Format Guidelines
 - **Quick Ref**: `C:\Users` → MCP: `C:\\Users` | Bash: `/c/Users` or `C:/Users`
-"""
+'''

.codex/agents/context-search-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "read-only"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are a context discovery specialist focused on gathering relevant project information for development tasks. Execute multi-layer discovery autonomously to build comprehensive context packages.
 
@@ -577,4 +577,4 @@ Output: .workflow/session/{session}/.process/context-package.json
 ### Windows Path Format Guidelines
 - **Quick Ref**: `C:\Users` → MCP: `C:\\Users` | Bash: `/c/Users` or `C:/Users`
 - **Context Package**: Use project-relative paths (e.g., `src/auth/service.ts`)
-"""
+'''

.codex/agents/debug-explore-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are an intelligent debugging specialist that autonomously diagnoses bugs through evidence-based hypothesis testing and CLI-assisted analysis.
 
@@ -434,4 +434,4 @@ ${nextSteps}
 - Timeout: Analysis 20min | Fix implementation 40min
 
 ---
-"""
+'''

.codex/agents/doc-generator.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 You are an expert technical documentation specialist. Your responsibility is to autonomously **execute** documentation tasks based on a provided task JSON file. You follow `flow_control` instructions precisely, synthesize context, generate or execute documentation generation, and report completion. You do not make planning decisions.
 
@@ -322,4 +322,4 @@ Before completing the task, you must verify the following:
 - **Generate Code**: Your role is to document, not to implement.
 - **Skip Quality Checks**: Always perform the full QA checklist before completing a task.
 - **Mix Modes**: Do not generate content in CLI Mode or execute CLI in Agent Mode - respect the `cli_execute` flag.
-"""
+'''

.codex/agents/issue-queue-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 ## Overview
 
@@ -309,4 +309,4 @@ Return brief summaries; full conflict details in separate files:
 ```
 - `clarifications`: Only present if unresolved high-severity conflicts exist
 - No markdown, no prose - PURE JSON only
-"""
+'''

.codex/agents/test-action-planning-agent.toml

@@ -4,7 +4,7 @@ model = "gpt-5.4"
 model_reasoning_effort = "high"
 sandbox_mode = "workspace-write"
 
-developer_instructions = """
+developer_instructions = '''
 
 ## Agent Inheritance
 
@@ -673,4 +673,4 @@ Hard Constraints:
 - AI issue detection configured in IMPL-001.3
 - Quality gates with measurable thresholds in IMPL-001.5
 - Source session status reported (if applicable)
-"""
+'''