What does Flow Prime assess?

It assesses eight pillars (30 agent-readiness criteria + 18 production-readiness criteria) to provide full visibility and targeted fixes for agent readiness.

Will Flow Prime modify my code?

No. It never modifies code files; fixes are offered for agent readiness and can be applied by you through remediation prompts.

How are scores and maturity levels determined?

Scores are computed from pillar results and mapped to maturity levels (1-5), with Level 3 typically the target for most teams. Production-readiness issues are reported but not fixed.

flow-next-prime

npx machina-cli add skill gmickel/gmickel-claude-marketplace/flow-next-prime --openclaw

Files (1)

SKILL.md

5.0 KB

Flow Prime

Comprehensive codebase assessment inspired by Factory.ai's Agent Readiness framework.

Role: readiness assessor, improvement proposer Goal: full visibility into codebase health, targeted fixes for agent readiness

Two-Tier Assessment

Category	Pillars	What Happens
Agent Readiness	1-5 (30 criteria)	Scored, maturity level calculated, fixes offered
Production Readiness	6-8 (18 criteria)	Reported for awareness, no fixes offered

This gives you full visibility while keeping remediation focused on what actually helps agents work.

Why This Matters

Agents waste cycles when:

No pre-commit hooks → waits 10min for CI instead of 5sec local feedback
Undocumented env vars → guesses, fails, guesses again
No CLAUDE.md → doesn't know project conventions
Missing test commands → can't verify changes work

These are environment problems, not agent problems. Prime helps fix them.

Input

Full request: $ARGUMENTS

Accepts:

No arguments (scans current repo)
--report-only or report only (skip remediation, just show report)
--fix-all or fix all (apply all agent readiness fixes without asking)
Path to different repo root

Examples:

/flow-next:prime
/flow-next:prime --report-only
/flow-next:prime ~/other-project

The Eight Pillars

Agent Readiness (Pillars 1-5) — Fixes Offered

Pillar	What It Checks
1. Style & Validation	Linters, formatters, type checking, pre-commit hooks
2. Build System	Build tools, commands, lock files, monorepo tooling
3. Testing	Test framework, commands, coverage, verification
4. Documentation	README, CLAUDE.md, setup docs, architecture
5. Dev Environment	.env.example, Docker, devcontainer, runtime version

Production Readiness (Pillars 6-8) — Report Only

Pillar	What It Checks
6. Observability	Logging, tracing, metrics, error tracking, health endpoints
7. Security	Branch protection, secret scanning, CODEOWNERS, Dependabot
8. Workflow & Process	CI/CD, PR templates, issue templates, release automation

Workflow

Read workflow.md and execute each phase in order.

Key phases:

Parallel Assessment — 9 haiku scouts run in parallel (~15-20 seconds)
Verification — Verify test commands actually work
Score & Synthesize — Calculate scores, determine maturity level
Present Report — Full report with all 8 pillars
Interactive Remediation — AskUserQuestion for agent readiness fixes only
Apply Fixes — Create/modify files based on selections
Summary — Show what was changed

Maturity Levels (Agent Readiness)

Level	Name	Description	Score
1	Minimal	Basic project structure only	<30%
2	Functional	Can build and run, limited docs	30-49%
3	Standardized	Agent-ready for routine work	50-69%
4	Optimized	Fast feedback loops, comprehensive docs	70-84%
5	Autonomous	Full autonomous operation capable	85%+

Level 3 is the target for most teams. Don't over-engineer.

What Gets Fixed vs Reported

Pillars	Category	Remediation
1-5	Agent Readiness	✅ Fixes offered via AskUserQuestion
6-8	Production Readiness	❌ Reported only, address independently

Guardrails

General

Never modify code files (only config, docs, scripts)
Never commit changes (leave for user to review)
Never delete files
Respect .gitignore patterns

User Consent

MUST use AskUserQuestion tool for consent — never just print questions as text
Always ask before modifying existing files
Don't add dependencies without consent

Scope Control

Never create LICENSE files — license choice requires explicit user decision
Never offer Pillar 6-8 fixes — production readiness is informational only
Focus fixes on what helps agents work (not team governance)

Scouts

Agent Readiness (haiku, fast)

tooling-scout — linters, formatters, pre-commit, type checking
claude-md-scout — CLAUDE.md/AGENTS.md analysis
env-scout — environment setup
testing-scout — test infrastructure
build-scout — build system
docs-gap-scout — README, ADRs, architecture

Production Readiness (haiku, fast)

observability-scout — logging, tracing, metrics, health
security-scout — GitHub settings, CODEOWNERS, secrets
workflow-scout — CI/CD, templates, automation

All 9 scouts run in parallel for speed.

Source

git clone https://github.com/gmickel/gmickel-claude-marketplace/blob/main/plugins/flow-next/skills/flow-next-prime/SKILL.md

View on GitHub

Overview

Flow Prime performs a comprehensive assessment across eight pillars (30 agent-readiness criteria and 18 production-readiness criteria) to reveal codebase health. It scans for issues like missing pre-commit hooks, undocumented env vars, and CLAUDE.md, verifies commands work, and checks GitHub settings. It reports all findings and applies fixes only for agent readiness.

How This Skill Works

On invocation, Flow Prime runs parallel assessments across the eight pillars, validates test commands, and computes maturity scores. It then presents a full eight-pillar report and, unless told otherwise, performs interactive remediation to fix agent-readiness gaps. Production-readiness issues are reported only and not auto-fixed.

When to Use It

You need a full health check across Agent Readiness pillars 1-5 and Production Readiness pillars 6-8.
You want to verify commands actually work in the repo and identify missing CI hooks or env vars.
You want a comprehensive report of eight pillars to inform remediation planning.
You want production-readiness issues reported only, with no changes applied.
You want to scan the current repo or point Flow Prime at a different repo root.

Quick Start

Step 1: Run /flow-next:prime in the repo root.
Step 2: Review the full eight-pillar report and note agent-readiness gaps.
Step 3: If fixes are desired, proceed with interactive remediation or use --fix-all.

Best Practices

Run Flow Prime from the repository root to scan the current repository by default.
Use --report-only to preview findings without applying fixes.
Review remediation prompts (AskUserQuestion) before applying any fixes.
Coordinate with developers; Flow Prime never modifies code files directly.
Ensure CLAUDE.md exists and environment variables are well-documented as part of remediation.

Example Use Cases

/flow-next:prime
/flow-next:prime --report-only
/flow-next:prime --fix-all
/flow-next:prime ~/enterprise-app
/flow-next:prime --report-only ~/other-repo

Frequently Asked Questions

Add this skill to your agents