using-forge

A guide for using the Forge coding-agent plugin, which adds product-planning steps around coding work. It explains two paths: one for testing a product idea before building it and one for planning and shipping an existing product.

In plain words
What is it for?
Choosing Founder or PM mode, checking which project file determines the mode, starting Forge when neither mode is set, and understanding the 14 supported commands.
Why use it?
It helps new Forge users choose the right starting point, understand its commands, and avoid asking the coding agent to invent product direction on its own.

Skill for Claude CodeCodex

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 skills/dormstern/forge/using-forge
Any agent
npx skills add dormstern/forge --skill using-forge
Clone the repo
git clone --depth 1 https://github.com/dormstern/forge

Made for: Claude Code, Codex.

Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,568 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 $0.00072 $0.01568
Opus 5 $0.00036 $0.00784
Sonnet 5 $0.00014 $0.00314
Haiku 4.5 $0.00007 $0.00157

Measured yesterday against content hash 4fd9fad57431, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

using-forge 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.

skills/using-forge/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Using Forge

Stop shipping features your AI agent invented.

Forge wraps Claude Code with the product discipline AI coding tools forgot to ship: a Devil's Advocate that attacks every plan before code is written, a mandatory 10% scope kill every release, and drift audits every 5 features. So Claude ships the right features — not all the features.

Two modes:

  • Founder mode — pre-PMF. SCOUT → INTERROGATE → SYNTHESIZE → GRADUATE. Falsifies hypotheses until you've earned the right to build.
  • PM mode — post-PMF (the main thing). HARVEST → SHAPE → BUILD → SHIP. Each cycle proposes a release plan, kills 10% of scope, runs Devil's Advocate, audits drift every 5 features.

You're in Founder mode if THESIS.md (or THESIS-minimal.md / THESIS-full.md) exists. PM mode if IDENTITY.md exists. Neither → run /forge:init (PM default for repos with package.json) or /forge:init --founder.

⚠ Canonical command vocabulary — these are the ONLY 14

When you recommend a Forge command to the user, it MUST come from this exact list. Do not invent variants. Do not abbreviate. Do not construct command names from agent names.

Command Mode What it does
/forge:try Both ★ 30-second magic-moment demo. Zero setup. The recommended starting point.
/forge:init Both Scaffold state files (PM default for repos with package.json; --founder for pre-PMF; --full for full Founder template)
/forge:cycle Both Run a full cycle, auto-routed by mode
/forge:status Both One-screen state snapshot with contextual interpretation
/forge:help Both Orientation — modes, commands, state files. Start here if confused.
/forge:falsify Both Standalone Devil's Advocate against any plan/pitch/PRD/thesis. (NOT /forge:advocate, /forge:da, /forge:devil — those don't exist.)
/forge:scout Founder Market scout — community mining
/forge:interrogate Founder Interview architect + mandatory DA
/forge:synthesize Founder Thesis synthesis with 1–5 confidence rubric
/forge:graduate Founder Hand off to PM mode when 12 criteria met
/forge:harvest PM Collect signals + refresh architecture
/forge:shape PM Hero command — propose release plan with kill list + DA
/forge:build PM Ralph Loop with autonomy tiers + drift checks
/forge:ship PM Log release + measure + check stabilization
/forge:drift PM Standalone drift audit on last 5 features

Read the full file on GitHub · 86 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 · 86 lines · 72 tokens per session scan A 4fd9fad57431

Subscribe to this mod's changes

using-forge is a skill published in the GitHub repository dormstern/forge (6 stars, last pushed 3mo ago), licensed MIT. It adds 72 tokens to every session and 1,568 once invoked, about $0.0004 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens