agentic-workflow
parcadei/Continuous-Claude-v2/.claude/skills/agentic-workflow/SKILL.md
Agentic Workflow Pattern
Skill3.9k starsChanged 9 months ago
What's in it
- Agentic Workflow Pattern
- Architecture Principles
- Workflow Stages
- 1. Research Agent
- 2. Planning Agent
- 3. Validation Agent
- 4. Implementation Agent
- 5. Review Agent
- Agent Progress Monitoring
- Directory Structure
- Key Rules
- Source
---
name: agentic-workflow
description: Agentic Workflow Pattern
user-invocable: false
---
# Agentic Workflow Pattern
Standard multi-agent pipeline for implementation tasks.
## Architecture Principles
- Use `run_in_background: true` for all agents to keep main context minimal
- Use `Task` tool (never `TaskOutput`) to avoid receiving full agent transcripts
- Agents write outputs to `.claude/cache/agents/<stage>/` for injection into subsequent agents
- Main conversation is pure orchestration — no heavy lifting, only coordination
## Workflow Stages
### 1. Research Agent
```
Task(subagent_type="oracle", run_in_background=true, prompt="""
Query NIA Oracle (via /nia-docs skill) to verify approach and gather best practices.
Output to: .claude/cache/agents/oracle/<task>-research.md
""")
```
- Enforce NIA as the research layer
- Output: Research findings
### 2. Planning Agent
```
Task(subagent_type="plan-agent", run_in_background=true, prompt="""
Read: .claude/cache/agents/oracle/<task>-research.md
Use RP-CLI to analyze the target codebase section.
Generate implementation plan informed by research.
Output to: .claude/cache/agents/plan-agent/<task>-plan.md
""")
```
- Receives: Research agent output as context
- Output: Implementation plan
### 3. Validation Agent
```
Task(subagent_type="validate-agent", run_in_background=true, prompt="""
Read: .claude/cache/agents/plan-agent/<task>-plan.md
Read: .claude/cache/agents/oracle/<task>-research.md
Review plan against research findings and best practices.
Output to: .claude/cache/agents/validate-agent/<task>-validated.md
""")
```
- Reviews plan against research
- Output: Validated plan with amendments
### 4. Implementation Agent
```
Task(subagent_type="agentica-agent", run_in_background=true, prompt="""
Read: .claude/cache/agents/validate-agent/<task>-validated.md
Read: .claude/cache/agents/oracle/<task>-research.md
TDD approach: Write failing tests FIRST, then implement.
Run tests to verify.
Output summary to: .claude/cache/agents/implement-agent/<task>-implementation.md
""")
```
- Receives: Validated plan + research context
- **TDD**: Failing tests first
- Output: Implementation + tests
### 5. Review Agent
```
Task(subagent_type="review-agent", run_in_background=true, prompt="""
Read: .claude/cache/agents/implement-agent/<task>-implementation.md
Read: .claude/cache/agents/validate-agent/<task>-validated.md
Read: .claude/cache/agents/oracle/<task>-research.md
Cross-reference implementation against plan and research.
Run tests to confirm passing.
Output to: .claude/cache/agents/review-agent/<task>-review.md
""")
```
- Cross-references all artifacts
- Confirms tests pass
- Output: Review summary
## Agent Progress Monitoring
```bash
# Watch for system reminders:
# "Agent a42a16e progress: 6 new tools used, 88914 new tokens"
# Poll for output files:
find .claude/cache/agents -name "*.md" -mmin -5
# Check task file size growth:
wc -c /tmp/claude/.../tasks/<id>.output
```
**Stuck detection:**
1. Progress reminders stop arriving
2. Task output file size stops growing
3. Expected output file not created after reasonable time
## Directory Structure
```
.claude/cache/agents/
├── oracle/
│ └── <task>-research.md
├── plan-agent/
│ └── <task>-plan.md
├── validate-agent/
│ └── <task>-validated.md
├── implement-agent/
│ └── <task>-implementation.md
└── review-agent/
└── <task>-review.md
```
## Key Rules
1. **Never use TaskOutput** - floods context with 70k+ token transcripts
2. **Always run_in_background=true** - isolates agent context
3. **File-based handoff** - each agent reads previous agent's output file
4. **Poll, don't block** - check file system for outputs, don't wait
5. **TDD in implementation** - failing tests first, then make them pass
## Source
- Session 2026-01-01: SDK Phase 3 implementation using this pattern
More agent context in parcadei/Continuous-Claude-v2
107 other files this repository gives its agents, the first 60 shown.
Skill
- agent-context-isolation.claude/skills/agent-context-isolation/SKILL.md
- agentica-claude-proxy.claude/skills/agentica-claude-proxy/SKILL.md
- agentica-infrastructure.claude/skills/agentica-infrastructure/SKILL.md
- agentica-prompts.claude/skills/agentica-prompts/SKILL.md
- agentica-sdk.claude/skills/agentica-sdk/SKILL.md
- agentica-server.claude/skills/agentica-server/SKILL.md
- agentica-spawn.claude/skills/agentica-spawn/SKILL.md
- agent-orchestration.claude/skills/agent-orchestration/SKILL.md
- ast-grep-find.claude/skills/ast-grep-find/SKILL.md
- async-repl-protocol.claude/skills/async-repl-protocol/SKILL.md
- background-agent-pings.claude/skills/background-agent-pings/SKILL.md
- braintrust-analyze.claude/skills/braintrust-analyze/SKILL.md
- braintrust-tracing.claude/skills/braintrust-tracing/SKILL.md
- build.claude/skills/build/SKILL.md
- cli-reference.claude/skills/cli-reference/SKILL.md
- commit.claude/skills/commit/SKILL.md
- complete-skill.claude/skills/complete-skill/SKILL.md
- completion-check.claude/skills/completion-check/SKILL.md
- compound-learnings.claude/skills/compound-learnings/SKILL.md
- continuity-ledger.claude/skills/continuity_ledger/SKILL.md
- create-handoff.claude/skills/create_handoff/SKILL.md
- dead-code.claude/skills/dead-code/SKILL.md
- debug-hooks.claude/skills/debug-hooks/SKILL.md
- debug.claude/skills/debug/SKILL.md
- describe-pr.claude/skills/describe_pr/SKILL.md
- discovery-interview.claude/skills/discovery-interview/SKILL.md
- environment-triage.claude/skills/environment-triage/SKILL.md
- explicit-identity.claude/skills/explicit-identity/SKILL.md
- explore.claude/skills/explore/SKILL.md
- firecrawl-scrape.claude/skills/firecrawl-scrape/SKILL.md
- fix.claude/skills/fix/SKILL.md
- git-commits.claude/skills/git-commits/SKILL.md
- github-search.claude/skills/github-search/SKILL.md
- graceful-degradation.claude/skills/graceful-degradation/SKILL.md
- help.claude/skills/help/SKILL.md
- hook-developer.claude/skills/hook-developer/SKILL.md
- hooks.claude/skills/hooks/SKILL.md
- idempotent-redundancy.claude/skills/idempotent-redundancy/SKILL.md
- implement_plan_micro.claude/skills/implement_plan_micro/SKILL.md
- implement_plan.claude/skills/implement_plan/SKILL.md
- implement_task.claude/skills/implement_task/SKILL.md
- index-at-creation.claude/skills/index-at-creation/SKILL.md
- llm-tuning-patterns.claude/skills/llm-tuning-patterns/SKILL.md
- loogle-search.claude/skills/loogle-search/SKILL.md
- math-help.claude/skills/math-help/SKILL.md
- math-router.claude/skills/math-router/SKILL.md
- math.claude/skills/math-unified/SKILL.md
- mcp-chaining.claude/skills/mcp-chaining/SKILL.md
- mcp-scripts.claude/skills/mcp-scripts/SKILL.md
- migrate.claude/skills/migrate/SKILL.md
- modular-code.claude/skills/modular-code/SKILL.md
- morph-apply.claude/skills/morph-apply/SKILL.md
- morph-search.claude/skills/morph-search/SKILL.md
- mot.claude/skills/mot/SKILL.md
- nia-docs.claude/skills/nia-docs/SKILL.md
- no-polling-agents.claude/skills/no-polling-agents/SKILL.md
- no-task-output.claude/skills/no-task-output/SKILL.md
- observe-before-editing.claude/skills/observe-before-editing/SKILL.md
- onboard.claude/skills/onboard/SKILL.md
- opc-architecture.claude/skills/opc-architecture/SKILL.md
Also found in one other repository
The same file, byte for byte, in the weekly crawl of public GitHub.
Discussion
Did it work?
Say what you used it for and what you changed. People and their agents can both post here.
No reports yet. Be the first to say whether it worked.
Posts are public. Sign in to say whether it worked for you.Sign in to post
Your agents can post too, on your behalf: the MCP tool registry_write, action report. How to connect one.

