Every stage has a workflow. Every step is yours.
Instead of a blank prompt, Forge routes each request to a workflow: a set sequence of steps, with checkpoints, that runs the same way for every engineer. Only the steps a job needs run. These are the built-in defaults; open any card to see its steps.
You don’t have to pick from this list.
This page is the reference, not the price of entry. With the plugin installed, say what you’re working on in your own words and Forge matches it to the workflows your organization turned on, at the start of a session or partway through. Naming a workflow still works.
A built-in workflow for every stage.
These workflows ship with Forge — each one encodes industry best practices for its stage of the software development lifecycle. Use them as they come, or tailor every step, gate, and skill to the way your team works.
Shape and scope the work before any code is written — discovery, definition, PRDs, and readiness.
Steps
- Check for existing work items
- Gather background context
- Clarify requirements
- Collect non-functional requirements
- Write the problem statement
- Score impact (RICE)
- Preview & create the tracker work item
- Generate the PRD
- Build the tech handoff plan
Map the system and shape the technical approach for an existing work item.
Steps
- Resolve the work item
- Scan the system architecture
- Interactive architecture discussion — record decisions
Steps
- Locate any existing atlas
- Discover the estate across repos and services
- Synthesize the findings into one picture
- Validate the reconstruction
- Publish the atlas
- Render the visual map
Turn a ready work item into shipped code, wave by wave.
Steps
- Check for prior planning on the ticket
- Fetch the work item
- Readiness check
- Analyze change impact (blast radius)
- Technical discovery
- Define the test strategy
- Write the implementation proposal
- Notify the engineering lead
Steps
- Receive the engineering handoff
- Readiness check
- Break the work into stories
- Plan dependencies & implementation waves
- Analyze change impact (blast radius)
- Technical discovery
- Define the test strategy
- Implement wave by wave
- Review the PR
- Complete & summarize delivery
Verify quality before and after the change lands — test strategy and structured PR review.
Steps
- Fetch the work item
- Readiness check
- Check AC quality & testability
- Review stories (INVEST)
- Analyze change impact
- Technical discovery
- Define the test strategy
Steps
- Fetch the work item
- Check branch freshness
- AI-augmented PR review — analysis, AC traceability, OWASP scan
- Customer impact analysis
- Surface manual deploy tasks
Work out what's shipping, then ship it — gated, versioned, and announced.
Steps
- Resolve the release scope
- Run pre-release gates — CI, blockers, migrations
- Draft versioned release notes
Steps
- Resolve the release scope
- Run pre-release gates
- Check branch freshness
- Reconcile traceability to tracked work
- Build the release notes
- Plan the rollout
- Plan observability
- Cut the release
- Plan feature exposure
- Publish & announce
Capture and resolve defects end to end.
Steps
- Capture the bug — severity, area, reproduction
- Preview & create the ticket
- Hand off to engineering
- Implement the fix (optional)
- Review the PR & surface deploy tasks (optional)
Stay on top of work already in flight — status, documentation, and edits.
Steps
- Fetch the work item
- Report status — progress, acceptance criteria, PR health
Steps
- Reconstruct the work item from the branch
- Offer to create or link a tracked item
Steps
- Fetch the work item
- Fetch tracker and PR comments
- Filter out ones already handled
- Discuss each remaining comment to an agreed recommendation
Shorter or narrower paths that share steps with the workflows above. Each is a real built-in with its own prompt.
Every step, gate, and skill — yours to configure.
Every workflow is a list of steps you control. Each step carries a model tier, an approval gate, and the skills it must run — set once for the whole org, or forked for a single team.
One feature, start to finish.
Here's how a team uses Forge workflows to ship one complex feature — from a fuzzy idea to merged code — each person reaching for the right workflow, whether it's a Forge default or a version their team has customized.
Pressure-tests the idea before anyone commits — frames the problem, checks roadmap fit, and scores it so only work worth doing moves forward.
Turns the validated idea into a scoped work item (PROJ-412) with a PRD and acceptance criteria, then hands off cleanly to the engineering lead.
Forge scans the codebase, then runs an interactive architecture discussion — surfacing the key technical questions and recording every decision before code begins.
Maps a test strategy to every acceptance criterion — unit, integration, and end-to-end coverage — while the code is still being written, not after.
Builds the feature wave by wave against the agreed plan — briefed on the architecture up front, so the first pass holds up in review — and opens a pull request.
A QA-team fork adds extra code-quality and test-coverage scans, pausing the merge when coverage slips.
A DevOps-team fork checks the change's impact on infrastructure performance and cost before it ships.
Pulls a read-only snapshot whenever it's needed — progress, acceptance criteria, and PR health, written straight back to Jira, Linear, or Notion — so there's no standup chase.
You don't have to start with Forge.
That journey reads top to bottom. Real sessions don't. Most work starts as a blank prompt — a bug someone's chasing, a refactor that grew, an idea still being thought out loud — and only turns into something worth tracking halfway through. Forge is built to join at that moment. Ask for a workflow at any point in a session and it picks up from where you already are: the conversation you've had, the files you've opened, and the decisions you've already made carry into it.
An engineer opens Claude Code and starts chasing a checkout bug. No ticket, no process — a blank prompt and forty minutes of digging through the codebase.
The root cause is a rounding helper three teams depend on. Suddenly this needs a ticket, a test strategy, and a review trail — and none of that exists yet.
It files the bug with what's already been found — the reproduction, the root cause, the files touched — then runs the fix as a workflow, with checkpoints, from there. Nothing gets re-explained, and the last forty turns land in the audit trail instead of evaporating.
Some workflows exist precisely for work already in flight — they assume a session, a branch, or a ticket that's been running for a while, and meet it where it is.
Reads the commits and diff on the branch you've been working on and reconstructs what you actually built — acceptance criteria mapped to real code and tests — then drafts the work item nobody wrote up front.
Pulls new comments from the ticket and its pull request, drops the ones already handled, and walks through the rest with you — built for the middle of an implementation, not the start of one.
A read-only snapshot whenever it's needed — progress, acceptance criteria, and PR health, written straight back to Jira, Linear, or Notion. No standup chase.
Runs the full structured review — code analysis, requirement traceability, security scan — on a PR that was opened without Forge ever being involved.
Works out what would ship right now from the delta since the last release, checks for blockers, and drafts the notes — without tagging or publishing anything.
Your process, your workflows.
Need something the defaults don't cover? An admin says so in Claude Code, Codex, or Cursor — “forge, create a workflow for our incident postmortems” — and describes it in plain English from there. Forge drafts it, shows exactly what changes, and saves it — across the org, or forked for a single team.
A forge-workflow session — plain-English request, a full keep/add plan, and an admin-confirmed save, scoped to one team.
What you can ask for
Just say what you want — no skill name to remember, no config files, no YAML. Each example shows something you could type (left) and the workflow Forge builds in response (right): extend a default, fork it for one team, or create a brand-new path.
A new capable-tier step that scans Terraform, Kubernetes & cloud-config diffs and surfaces cost impact before the PR can merge.
A QA-scoped fork of PR Review with an added coverage gate set to Always ask — merge pauses when coverage drops.
A fast-tier step that refreshes your CLAUDE.md / AGENTS.md so agent guidance never drifts from the code.
An Always-pause checkpoint wired to your security-review skill, gated to runs that touch authentication or PII.
A brand-new path — three ordered steps ending in a mandatory human approval gate before anything ships.
Find out which of these
your team needs.
Setup reads how your team already ships and turns on what closes the gaps — with a reason for each. Free for individuals. Install the plugin, then run guided setup.