epic-dev

A workflow for delivering a large feature in stages using Git worktrees, GitHub issues, and pull requests. A Git worktree is a separate working folder connected to the same repository, so multiple branches can be developed at once.

In plain words
What is it for?
Use it to plan an epic, create linked GitHub issues and sub-issues, develop phases in separate worktrees, implement and review tasks with coding agents, and deliver the results through pull requests.
Why use it?
Large changes need planning, progress tracking, review, and controlled delivery. This workflow divides an epic into smaller tasks and phases, assigns implementation and review work, and pauses for approval at key points.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/neuromechanist/research-skills/epic-dev
Clone the repo
git clone --depth 1 https://github.com/neuromechanist/research-skills
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,554 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00017 $0.02554
Opus 5 $0.00009 $0.01277
Sonnet 5 $0.00003 $0.00511
Haiku 4.5 $0.00002 $0.00255

Measured yesterday against content hash f558b41dbe7b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

epic-dev scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/project/commands/epic-dev.md · 238 lines

How it starts

The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Epic Development Workflow

Orchestrate multi-phase feature development using git worktrees, GitHub issues with sub-issues, and phased PR delivery. Load the project:workflow-reference skill immediately for reference context.

Core Principles

  • Full automation: Execute all git, gh, and worktree commands automatically
  • Confirm only at strategic points: epic plan, implementation plan, critical review findings, final merge
  • Use absolute paths: Bash calls do not persist cd; always use full worktree paths
  • Track progress: Use TodoWrite to track all phases and steps
  • No emojis: Never use emojis in commits, PRs, or issues
  • Model routing: Keep epic design, observation, approvals, and synthesis on Fable when available, otherwise Opus. Delegate approved-plan implementation, focused review, and validation to named Sonnet agents. Escalate unresolved architecture and high-risk invariants back to the lead model.
  • GitHub prose: Semantic line breaks remain the prose-source default, but GitHub issue and PR bodies are the exception. Keep each paragraph on one source line, separate paragraphs with blank lines, and do not insert sentence- or clause-level newlines within a paragraph.
  • Agent cleanup: After incorporating a final report, close/remove the one-off agent unless a concrete follow-up task is already assigned.

Phase 0: Repository Detection

Goal: Detect repo configuration and validate prerequisites.

Actions:

  1. Run project-detect-repo-config and parse the output
  2. Validate all required values are present: INTEGRATION_BRANCH, REPO_ROOT, CURRENT_BRANCH must be non-empty. If any are missing or if ERROR is set, report the error to the user and stop.
  3. Validate gh is functional: If HAS_GH=false or GH_ERROR is set, tell the user what needs to be fixed and stop.
  4. If HAS_GH_SUBISSUE=false, install it: gh extension install agbiotech/gh-sub-issue
  5. Store detected values: INTEGRATION_BRANCH, REPO_ROOT, HAS_EPIC_STATE
  6. Create initial todo list tracking all workflow phases

Read the full file on GitHub · 238 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 238 lines · 0 tokens per session scan A f558b41dbe7b

Subscribe to this mod's changes

epic-dev is a command published in the GitHub repository neuromechanist/research-skills (43 stars, last pushed 2d ago), licensed BSD-3-Clause. It adds 17 tokens to every session and 2,554 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.