Need feedback on my agentic workflow for my team
retained text
I’ve been working on a structured Claude Code development workflow for my team of 100+ devs. We have a limited monthly token budget per developer, so I’m trying to keep things simple without relying on multiple subagents. Here’s the workflow: /new-story → Takes the user story and technical design, sets up the workspace. /plan-story → Analyzes the codebase, breaks the story into smaller tasks, and waits for developer approval. /build-task → Implements one task at a time, runs tests and validation, and updates progress. /story-status → Tracks completed, pending, and blocked tasks. Helps resume across sessions. /review-story → Reviews the implementation against requirements and design, identifies issues. /create-mr → Checks validation evidence and prepares the merge request for developer review. Alongside this, we also have AGENTS.md, skills, and hooks configured for project instructions, coding standards, security guardrails, and automated checks. **The only downside I see is that developers have to run each command manually.** I could automate the entire flow, but I’m unsure if the added complexity and token cost would be worth it. Would you automate this, keep it manual, or simplify the workflow? Am I missing anything important?
document fetched 2026-11-12 00:19 · canonical https://reddit.com/r/AI_Agents/comments/1xt74d5/need_feedback_on_my_agentic_workflow_for_my_team