digraph strategy_selection {
    rankdir=TB;
    "What type of work?" [shape=diamond];

    "Research / knowledge gathering" [shape=box];
    "Independent feature builds" [shape=box];
    "Sequential dependent tasks" [shape=box];
    "Same transformation across partitions" [shape=box];
    "Codebase audit / assessment" [shape=box];
    "Greenfield project kickoff" [shape=box];

    "Research Swarm" [shape=box style=filled fillcolor=lightyellow];
    "Epic Parallel Build" [shape=box style=filled fillcolor=lightyellow];
    "Sequential Pipeline" [shape=box style=filled fillcolor=lightyellow];
    "Parallel Sweep" [shape=box style=filled fillcolor=lightyellow];
    "Multi-Dimensional Audit" [shape=box style=filled fillcolor=lightyellow];
    "Full Lifecycle" [shape=box style=filled fillcolor=lightyellow];

    "What type of work?" -> "Research / knowledge gathering";
    "What type of work?" -> "Independent feature builds";
    "What type of work?" -> "Sequential dependent tasks";
    "What type of work?" -> "Same transformation across partitions";
    "What type of work?" -> "Codebase audit / assessment";
    "What type of work?" -> "Greenfield project kickoff";

    "Research / knowledge gathering" -> "Research Swarm";
    "Independent feature builds" -> "Epic Parallel Build";
    "Sequential dependent tasks" -> "Sequential Pipeline";
    "Same transformation across partitions" -> "Parallel Sweep";
    "Codebase audit / assessment" -> "Multi-Dimensional Audit";
    "Greenfield project kickoff" -> "Full Lifecycle";
}

Phase 1: Deploy research army (ALL BACKGROUND)
    Wave 1 (10-20 agents): Core technology research
    Wave 2 (10-20 agents): Specialized topics, integrations
    Wave 3 (5-10 agents): Gap-filling based on early results

Phase 2: Monitor and supplement
    - Check completed docs as they arrive
    - Identify gaps, deploy targeted follow-up agents
    - Read completed research to inform remaining dispatches

Phase 3: Synthesize
    - Read all research docs (foreground)
    - Create architecture plans, design docs
    - Use Plan agent to synthesize findings

Research [TECHNOLOGY] for [PROJECT]'s [USE CASE].

Create a comprehensive research doc at [OUTPUT_PATH]/[filename].md covering:

1. Latest [TECH] version and features (search "[TECH] 2026" or "[TECH] latest")
2. [Specific feature relevant to project]
3. [Another relevant feature]
4. [Integration patterns with other stack components]
5. [Performance characteristics]
6. [Known gotchas and limitations]
7. [Best practices for production use]
8. [Code examples for key patterns]

Include code examples where possible. Use WebSearch and WebFetch to get current docs.

Phase 1: Scout (FOREGROUND)
    - Deploy one Explore agent to map the codebase
    - Identify dependency chains and independent workstreams
    - Group tasks by subsystem to prevent file conflicts

Phase 2: Deploy build army (ALL BACKGROUND)
    Wave 1: Infrastructure/foundation (Redis, DB, auth)
    Wave 2: Backend APIs (each in own module directory)
    Wave 3: Frontend pages (each in own route directory)
    Wave 4: Integrations (MCP servers, external services)
    Wave 5: DevOps (CI, Docker, deployment)
    Wave 6: Bug fixes from review findings

Phase 3: Monitor and coordinate
    - Check git status for completed commits
    - Handle git index.lock contention (expected with 30+ agents)
    - Deploy remaining tasks as agents complete
    - Track via Sibyl tasks or TodoWrite

Phase 4: Review and harden (FOREGROUND)
    - Run Codex/code-reviewer on completed work
    - Dispatch fix agents for critical findings
    - Integration testing

**Task: [DESCRIPTIVE TITLE]** (task\_[ID])

Work in /path/to/project/[SPECIFIC_DIRECTORY]

**Key rules:**

- Every agent gets its own directory scope -- NO OVERLAP
- Provide existing patterns to follow ("Follow pattern from X")
- Include infrastructure context ("Redis available at X")
- Explicit git hygiene instructions (critical with 30+ parallel agents)
- Task IDs for traceability

For each task:
    1. Dispatch implementer (FOREGROUND)
    2. Dispatch spec reviewer (FOREGROUND)
    3. Dispatch code quality reviewer (FOREGROUND)
    4. Fix any issues found
    5. Move to next task

Trust Gradient (adapt over time):
    Early tasks:  Implement -> Spec Review -> Code Review (full ceremony)
    Middle tasks: Implement -> Spec Review (lighter)
    Late tasks:   Implement only (pattern proven, high confidence)

Phase 1: Analyze the scope
    - Run the tool (ruff, ty, etc.) to get full issue list
    - Auto-fix what you can
    - Group remaining issues by module/directory

Phase 2: Fan-out fix agents (4-10 agents)
    - One agent per module/directory
    - Each gets: issue count by category, domain-specific guidance
    - All foreground (need to verify each completes)

Phase 3: Verify and repeat
    - Run the tool again to check remaining issues
    - If issues remain, dispatch another wave
    - Repeat until clean

Fix all [TOOL] issues in the [MODULE_NAME] directory ([PATH]).

Current issues ([COUNT] total):

- [RULE_CODE]: [description] ([count]) -- [domain-specific fix guidance]
- [RULE_CODE]: [description] ([count]) -- [domain-specific fix guidance]

Run `[TOOL_COMMAND] [PATH]` to see exact issues.

IMPORTANT for [DOMAIN] code:
[Domain-specific guidance, e.g., "GTK imports need GI.require_version() before gi.repository imports"]

After fixing, run `[TOOL_COMMAND] [PATH]` to verify zero issues remain.

Dispatch 6 parallel reviewers (ALL FOREGROUND):
    1. Code quality & safety reviewer
    2. Integration correctness reviewer
    3. Spec completeness reviewer
    4. Test coverage reviewer
    5. Performance analyst
    6. Security auditor

Wait for all to complete, then:
    - Synthesize findings into prioritized action list
    - Dispatch targeted fix agents for critical issues
    - Re-review only the dimensions that had findings

[DIMENSION] review of [COMPONENT] implementation.

**Files to review:**

- [file1.ext]
- [file2.ext]
- [file3.ext]

**Analyze:**

1. [Specific question for this dimension]
2. [Specific question for this dimension]
3. [Specific question for this dimension]

**Report format:**

- Findings: numbered list with severity (Critical/Important/Minor)
- Assessment: Approved / Needs Changes
- Recommendations: prioritized action items

Session 1: RESEARCH (Research Swarm)
    -> 30-60 background agents build knowledge corpus
    -> Architecture planning agents synthesize findings
    -> Output: docs/research/*.md + docs/plans/*.md

Session 2: BUILD (Epic Parallel Build)
    -> Scout agent maps what exists
    -> 30-60 background agents build features by epic
    -> Monitor, handle git contention, track completions
    -> Output: working codebase with commits

Session 3: ITERATE (Build-Review-Fix Pipeline)
    -> Code review agents assess work
    -> Fix agents address findings
    -> Deep audit agents (foreground) assess each subsystem
    -> Output: quality-assessed codebase

Session 4: HARDEN (Sequential Pipeline)
    -> Integration boundary reviews (foreground, sequential)
    -> Security fixes, race condition fixes
    -> Test infrastructure setup
    -> Output: production-ready codebase

digraph bg_fg {
    "What is the agent producing?" [shape=diamond];

    "Information (research, docs)" [shape=box];
    "Code modifications" [shape=box];

    "Does orchestrator need it NOW?" [shape=diamond];
    "BACKGROUND" [shape=box style=filled fillcolor=lightgreen];
    "FOREGROUND" [shape=box style=filled fillcolor=lightyellow];

    "Does next task depend on this task's files?" [shape=diamond];
    "FOREGROUND (sequential)" [shape=box style=filled fillcolor=lightyellow];
    "FOREGROUND (parallel)" [shape=box style=filled fillcolor=lightyellow];

    "What is the agent producing?" -> "Information (research, docs)";
    "What is the agent producing?" -> "Code modifications";

    "Information (research, docs)" -> "Does orchestrator need it NOW?";
    "Does orchestrator need it NOW?" -> "FOREGROUND" [label="yes"];
    "Does orchestrator need it NOW?" -> "BACKGROUND" [label="no - synthesize later"];

    "Code modifications" -> "Does next task depend on this task's files?";
    "Does next task depend on this task's files?" -> "FOREGROUND (sequential)" [label="yes"];
    "Does next task depend on this task's files?" -> "FOREGROUND (parallel)" [label="no - different modules"];
}

You are researching [DOMAIN] to create comprehensive documentation for [PROJECT].

Your mission: Create an exhaustive reference document covering ALL [TOPIC] capabilities.

Cover these areas in depth:

1. **[Category]** -- specific items
2. **[Category]** -- specific items
   ...

Use WebSearch and WebFetch to find blog posts, GitHub repos, and official docs.

**Task: [TITLE]** (task\_[ID])

Work in /absolute/path/to/[directory]

undefined

Comprehensive audit of [SCOPE] for [DIMENSION].

Look for:

1. [Specific thing #1]
2. [Specific thing #2]
   ...
3. [Specific thing #10]

[Scope boundaries -- which directories/files]

Report format:

- Findings: numbered with severity
- Assessment: Pass / Needs Work
- Action items: prioritized

**Task:** Fix [ISSUE] -- [SEVERITY]

**Problem:** [Description with file:line references]
**Location:** [Exact file path]

**Fix Required:**

1. [Specific change]
2. [Specific change]

**Verify:**

1. Run [command] to confirm fix
2. Run tests: [test command]

undefined

---