DevFlow Structured Development Workflow

VerifiedCaution

Progressive disclosure workflow with gate-based verification. Use for staged development where tests must pass before advancing.

Sby Skills Guide Bot
DevelopmentIntermediate
8/4/2026
Claude CodeCursorWindsurfCopilotCodex
#workflow#gated-verification#cli#feature-development#debugging

Recommended for

Automated summary

DevFlow is a CLI-driven progressive workflow in which the agent advances one verified step at a time, with each transition gated by explicit verification checks.

Strengths

  • Enforces step-by-step progression with gate checks (files, tests, approvals) before advancing
  • The agent-devflow pip package and .devflow/ directory make setup and initialization explicit and reproducible
  • Handles failures through fail routes and escalation to a human after three failed attempts
  • Cleanly separates selectable workflows (list-workflows / select-workflow) from current state (current)

Limitations

  • Requires installing the devflow CLI via pip, adding an external dependency to the project
  • The strict loop (current → execute → done) can slow down trivial, small tasks
  • Verification is limited to predefined gate types (file_exists, command_success, etc.), constraining customization
When to use it

Use it when a feature build or debugging session needs a structured, staged workflow where tests must pass before each step.

When not to use it

Avoid it for a simple one-off change that doesn't warrant a staged workflow.

Security analysis

Caution
Quality score82/100

The skill contains no destructive, exfiltrating, or obfuscated instructions. Its risk comes from normal-but-powerful operations: installing an external package, running a CLI, executing test commands via gates, and an optional autonomous loop. These are legitimate for a development workflow, so it warrants caution rather than a risky rating.

Findings
  • •Instructs installing a third-party PyPI package (`pip install agent-devflow`) and optionally `pip install -e .` from source, which is a supply-chain trust dependency.
  • •`devflow run` starts an autonomous 'Ralph Loop' that advances steps without per-step human confirmation; combined with `command_success` gates that execute arbitrary test commands, this grants broad execution authority.
  • •No allowed-tools are declared, so the agent's actual capabilities are unspecified while the skill assumes shell access.
How this analysis is made

Examples

Start a gated feature workflow
Initialize devflow for this Python project and start a structured feature workflow: run devflow init, pick a workflow, then show me the current step before writing any code.
Advance step by step
Run devflow current, do exactly what the current step asks, then run devflow done and report the gate result.
Systematic debugging
Use devflow to debug this failing test: set up a workflow, advance only through devflow done, and escalate to me if gates fail three times.

name: devflow description: Use when a task requires structured development workflow with gate-based verification, progressive disclosure, or step-by-step advancement. Use for feature development, debugging, or any task where tests must pass before proceeding to the next step.

DevFlow

DevFlow is a CLI-driven progressive disclosure workflow. You advance one verified step at a time.

When to Use

  • Task requires writing requirements, design docs, or code in stages
  • You need gate-based verification (tests must pass before advancing)
  • User mentions devflow commands or .devflow/ directory
  • Structured feature development or systematic debugging is needed

When NOT to use: The task is a simple one-off change that doesn't need staged workflow.

Quick Start

devflow --help          # Verify CLI is installed

If devflow is not found, install first:

pip install agent-devflow     # From PyPI

Or from source:

pip install -e .

If the project does not have a .devflow/ directory, initialize it:

devflow init --language <language> --name "<project-name>"

Then start the workflow:

devflow list-workflows              # See available workflows
devflow select-workflow MODE-A      # Pick one
devflow current                     # Read current step

Core Loop (ALWAYS follow this)

1. devflow current     → Read what to do
2. Execute the step    → Write code, create docs, run tests
3. devflow done        → Check gates → advance or retry
4. Repeat until "Workflow complete!"

Core rule: Only advance through devflow done. Never skip steps. Never create files for future steps.

Commands

| Command | Purpose | |---------|---------| | devflow current | Show current step instruction | | devflow done | Check gates and advance | | devflow back | Go back one step | | devflow approve ITEM | Mark item as user-approved | | devflow set KEY VALUE | Set state variable | | devflow list-workflows | List available workflows | | devflow select-workflow ID | Select and start a workflow | | devflow run | Start autonomous Ralph Loop |

Gate Types

Gates block advancement until satisfied:

  • file_exists:path — File must exist
  • file_contains:path:content — File must contain text
  • command_success:{test_command} — Command exits 0
  • user_approved:ITEM — Requires devflow approve ITEM
  • state_set:var — State variable must be set

Fail Routes

When gates fail, the workflow may route based on failure count:

fails 1-2 → retry or fallback step
fails 3+  → escalate to human

Fail count persists across visits (resets only when gates pass).

Common Mistakes

| Mistake | Fix | |---------|-----| | Writing code before devflow current | Stop. Run devflow current first | | Creating files for future steps | Only create what the current step asks for | | Skipping devflow done | All advancement must go through devflow done | | Ignoring gate failures | Fix the issue, then devflow done again |

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/devflow-structured-development-workflow-mse7ioj0' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/devflow-structured-development-workflow-mse7ioj0/download' --output '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md.part' && mv '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md.part' '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && test -s '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/devflow' && cp -R '.skill-downloads/devflow-structured-development-workflow-mse7ioj0' '.claude/skills/devflow'

Cursor
test -f '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && test -s '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/devflow' && cp -R '.skill-downloads/devflow-structured-development-workflow-mse7ioj0' '.cursor/skills/devflow'

Codex
test -f '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && test -s '.skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md' && mkdir -p '.agents/skills' && test ! -e '.agents/skills/devflow' && cp -R '.skill-downloads/devflow-structured-development-workflow-mse7ioj0' '.agents/skills/devflow'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/devflow-structured-development-workflow-mse7ioj0'

openskills read 'devflow-structured-development-workflow-mse7ioj0'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/devflow-structured-development-workflow-mse7ioj0/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills