agentleFS
pankaj-mahaurpankaj-mahaur/Agent-Daemon/skills/daemon/feature-flow/SKILL.md

feature-flow

Skill0 starsChanged 6 days agoMITView on GitHub

Use when the user wants a feature built end to end — \"build the whole feature\", \"feature start to finish\", \"end to end implement\", \"poora feature banao\", \"plan + build + verify\", \"ship this feature properly\". Chains feature-prep-docs → implement-feature → verify → review-slice as one disciplined workflow with checkpoints between steps.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a step's skill is not installed, say so and offer ad skill install <name> (or proceed with the step's intent manually, flagged as …

    line 19
  2. Distinctive:

    Never skip a checkpoint: if a step ended without one, re-state it before moving on (drift between steps is the #1 flow failure mode).

    line 50

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Feature Flow

A composite workflow: each step invokes an installed skill, ends with a one-line checkpoint, and the next step starts by validating that checkpoint. Steps are ordinary skill invocations — telemetry and GEPA evolution track each sub-skill individually, and this flow body itself is GEPA-evolvable.

If a step's skill is not installed, say so and offer ad skill install <name> (or proceed with the step's intent manually, flagged as "unskilled").

Steps

Step 1 — Plan

  • uses: feature-prep-docs
  • entry: a feature request with enough scope to size (ask 2-3 questions if not)
  • exit checkpoint: write to the conversation — FLOW feature-flow 1/4 planned: <doc paths>

Step 2 — Implement

  • uses: implement-feature
  • entry: Step 1's docs exist and the user approved the plan
  • discipline: search existing utilities before writing; one commit per root cause
  • exit checkpoint: FLOW feature-flow 2/4 implemented: <files touched>

Step 3 — Verify

  • uses: verify
  • entry: implementation compiles / lints clean
  • discipline: run the actual app or tests and OBSERVE behavior — type-check passing is not "tested" (constitution rule 1)
  • exit checkpoint: FLOW feature-flow 3/4 verified: <what was observed>

Step 4 — Review

  • uses: review-slice
  • entry: verification passed
  • discipline: findings grouped by root cause + severity; fix approved findings
  • exit checkpoint: FLOW feature-flow 4/4 reviewed: <findings count, fixes applied>

Rules

  • Never skip a checkpoint: if a step ended without one, re-state it before moving on (drift between steps is the #1 flow failure mode).
  • A failed step does NOT advance — fix or surface to the user.
  • Mid-flow scope changes restart at Step 1 (re-plan), not Step 2.
  • Record one activeContext.md line at flow completion: - YYYY-MM-DD: feature-flow completed for <feature> (<commit range>)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.