forge

A single entry point for a software-development workflow that turns a product idea into a plan and then runs it. It can reuse existing product requirements, specifications, or plans.

In plain words
What is it for?
Use it to design product requirements, optionally define technical details, create an implementation plan, estimate token costs, and execute approved work.
Why use it?
It keeps the workflow's review and approval points in place while avoiding repeated work when documents already exist.

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/alanvaa06/forge-master/forge
Any agent
npx skills add alanvaa06/forge-master --skill forge
Clone the repo
git clone --depth 1 https://github.com/alanvaa06/forge-master

Made for: Claude Code, Codex.

Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 881 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 $0.00092 $0.00881
Opus 5 $0.00046 $0.00441
Sonnet 5 $0.00018 $0.00176
Haiku 4.5 $0.00009 $0.00088

Measured yesterday against content hash da31666b23ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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/forge/SKILL.md · 40 lines

How it starts

The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.

forge — Full Pipeline (Umbrella)

You chain the four stages below in order. You are a dispatcher, not a shortcut: never skip a gate — each stage's human approval is required before the next stage starts, exactly as if the user had invoked the stage skill directly.

Stage 0 — Detect existing artifacts (enter at the right stage)

Before starting from scratch, list what already exists and ask the user where to enter:

  • docs/forge/prd/NNN-*.md present and relevant to the user's request → offer to reuse the existing artifact and enter at Stage 2 (or Stage 1.5) instead of redoing the PRD. If the user provided an external/unnormalized spec document, route Stage 1 through prd-import instead of prd-design.
  • docs/forge/specs/spec-NNN.md present for that PRD → offer to enter at Stage 2.
  • docs/forge/plans/plan-NNN.md present and approved → offer to enter at Stage 3.
  • Partial todo.md for that plan → this is a resume; go straight to Stage 3 (run INIT handles resume detection).

Present the entry point as a lettered list of only the stages that apply given what you found, mark one Recommended with a concrete why (the furthest-along stage matching the request — reusing approved artifacts beats redoing them), and let the user pick by letter. Never assume the entry stage. Example:

Found docs/forge/prd/002-billing.md (approved) and spec-002.md. Enter at: a) Stage 2 — plan-design using the existing PRD + spec — Recommended, both artifacts already approved b) Stage 1.5 — revise the spec first c) Stage 1 — start a fresh PRD (ignores existing)

Stage 1 — PRD (human gate 1)

Invoke the prd-design skill (or prd-import for an existing document) and follow it fully — divergent exploration, convergent interview, hard AC rules. Stop at its approval gate. Do not continue until the user explicitly approves the PRD.

Stage 1.5 — Spec (optional, human gate 1.5)

After PRD approval, ask the spec-or-plan question exactly as prd-design/prd-import specify — a lettered list with one option marked Recommended, user decides. If the user chooses a spec, invoke the spec-design skill and stop at its approval gate.

Read the full file on GitHub · 40 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 · 40 lines · 92 tokens per session scan A da31666b23ac

Subscribe to this mod's changes

forge is a skill published in the GitHub repository alanvaa06/forge-master (1 stars, last pushed 1mo ago), licensed MIT. It adds 92 tokens to every session and 881 once invoked, about $0.0005 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

goal-flight

Portable Goal Flight workflow for long-running repo work: planning, dispatch, review, recovery, file-backed resume.

simonrowland/goal-flight · 25 tokens

gossipcat

Use when installing or setting up gossipcat multi-agent orchestration (parallel review, consensus, adaptive dispatch) in Claude Code or Cursor. Installs the gossipcat MCP server and hands off to gossipstatus() for all live rules.

gossipcat-ai/gossipcat-ai · 50 tokens

code-review

Orchestrates code review by detecting the project language and architecture, then routing to the appropriate specialized review skill (code-review-go, code-review-typescript, code-review-python, code-review-php, code-review-architecture). Falls back to the generic checklist when no specific skill applies. Invoked when…

soulcodex/agentic · 80 tokens

new-gh-issue-orchestration

Orchestrates a GitHub-issue-driven delivery workflow from issue intake to PR creation using reviewer-first then worker execution. Invoked when the user provides a GitHub issue link/number and asks to start end-to-end delivery.

soulcodex/agentic · 54 tokens

surge

Use when a user provides a PRD, spec, or detailed requirements document and needs a full project delivered through iterative expert orchestration — multi-round analyze/research/design/implement/QA cycles with convergence detection. NOT for: single-file edits, quick prototypes, simple Q&A, or tasks without a written…

carbonshow/intent-fluid · 66 tokens

agentic-infrastructure-2026

Build and adopt production AI agent infrastructure in 2026. Covers framework selection (LangGraph, CrewAI, AutoGen, MCP), orchestration patterns, evaluation, observability, memory systems, and tool use. Also covers the SOCIAL dimension: how to sell agent infrastructure internally, change management, measuring ROI…

curiositech/windags-skills · 173 tokens