Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Agentic Workflow Pattern

A standardized five-stage multi-agent pipeline — research, plan, validate, TDD implement, review — using background agents and file-based handoff.

AutomationAdvanced★ 3,931⑂ 299AI score 5/10Last updated: Jan 26, 2026

What it does

  • Breaks an implementation task into a research → plan → validate → implement → review pipeline, one sub-agent per stage.
  • Enforces run_in_background: true and bans TaskOutput so 70k+ token agent transcripts never flood the main context.
  • Uses file-based handoff: each agent writes to .claude/cache/agents/<stage>/<task>-*.md and the next agent reads that file.
  • Provides progress polling recipes (find ... -mmin -5, output file growth) and stuck-agent detection heuristics.
  • Mandates TDD in the implementation stage: failing tests first, then make them pass.

Who it's for

  • Developers already orchestrating multiple Claude Code sub-agents on sizable features.
  • Anyone whose long sessions blow up the main conversation context.
  • Teams that have (or will create) custom agents named oracle, plan-agent, validate-agent, agentica-agent, review-agent.

Examples

  1. "Add webhook retry logic to the payments module" → oracle writes a research file, plan-agent reads it and drafts a plan, validate-agent audits it, then the implementation agent codes test-first.
  2. Large refactor where you want a lean main thread → only markdown artifacts persist; the main conversation just coordinates.
  3. Suspect a background agent stalled → run find .claude/cache/agents -name "*.md" -mmin -5 to see whether new output appeared.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/parcadei/Continuous-Claude-v3/HEAD/.claude/skills/agentic-workflow/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .claude/skills/agentic-workflow folder from the GitHub repo parcadei/Continuous-Claude-v3 into my ~/.claude/skills/agentic-workflow/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/parcadei/Continuous-Claude-v3.git /tmp/cc-v3 && mkdir -p ~/.claude/skills && cp -r /tmp/cc-v3/.claude/skills/agentic-workflow ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and move to a scratch directory.
  2. Clone the repo: git clone https://github.com/parcadei/Continuous-Claude-v3.git
  3. Create the skills directory: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r Continuous-Claude-v3/.claude/skills/agentic-workflow ~/.claude/skills/
  5. The skill depends on sub-agents named oracle, plan-agent, validate-agent, agentica-agent, review-agent. Copy the repo's .claude/agents/ folder too, or edit SKILL.md to use agent names that exist in your setup.
  6. Optional: pre-create the cache dir with mkdir -p .claude/cache/agents.
  7. Restart Claude Code, then hand it an implementation task and say "follow the agentic-workflow pattern" to activate the pipeline.
View source on GitHub ↗License: MIT