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.
npx agentmods add skills/forgeyclap/claude-forge/forge-extensionnpx skills add ForgeyClap/claude-forge --skill forge-extensiongit clone --depth 1 https://github.com/ForgeyClap/claude-forgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/forgeyclap/claude-forge/forge-extension)<a href="https://agentmods.dev/skills/forgeyclap/claude-forge/forge-extension"><img src="https://agentmods.dev/badge/skills/forgeyclap/claude-forge/forge-extension.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00041 | $0.01295 |
| Opus 5 | $0.00020 | $0.00647 |
| Sonnet 5 | $0.00008 | $0.00259 |
| Haiku 4.5 | $0.00004 | $0.00129 |
Grade A, and why
forge-extension 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 5d ago.
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.
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 playbook — Browser extension (Chrome MV3)
build-boss leads; security-boss co-owns the permission + messaging + CSP surface (this is a security-heavy domain and a real Codex/security trigger). Popup/options UI defers to forge-website (real content, responsive, a11y); anything that calls a backend defers to forge-integration. Scope is Chrome/Edge Manifest V3 unless the owner asks for Firefox/Safari (those are separate ports). The honesty rule: an extension that "loads unpacked" must have its core flow actually exercised, and every permission requested must be one the code genuinely uses.
Hard rules (non-negotiable)
- Least-privilege permissions. Request only the
permissions+host_permissionsthe code actually uses; preferactiveTabandoptional_permissionsover a blanket<all_urls>. Each permission is justifiable in one sentence. Over-broad scope is both a Web Store rejection risk and a user-trust failure. - No remotely-hosted code (MV3 policy). All executable JS ships inside the package — no
eval,new Functionon remote strings, or<script src>from a CDN. Remote code is a hard Chrome Web Store rejection. - Safe content-script + messaging. Content scripts run in the page's hostile context: validate every
chrome.runtime.onMessagepayload and checksender; never trust orevalwindow.postMessagedata from the page; keepexternally_connectablenarrow.web_accessible_resourcesscoped to specific matches, never<all_urls>. - Strict CSP kept strict. MV3's default extension CSP (
script-src 'self') is not weakened — no'unsafe-eval', no'unsafe-inline', no remote script sources. Sandboxed pages only if genuinely required, narrowly. - Termination-resilient service worker. The MV3 background is an event-driven service worker, not a persistent page: listeners registered at top level, no reliance on in-memory global state surviving, durable state persisted to
chrome.storage. - No secrets in the package; store-policy compliance. No API secret baked into the extension (it's extractable); real secrets sit behind a backend, OAuth via
chrome.identitywith minimal scopes. Single clear purpose, accurate description, privacy disclosure if user data is handled, no obfuscated code.
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.
- 5d ago First seen · 40 lines · 41 tokens per session scan A 657f0fdd529f
forge-extension is a skill published in the GitHub repository ForgeyClap/claude-forge (2 stars, last pushed 1mo ago), licensed MIT. It adds 41 tokens to every session and 1,295 once invoked, about $0.0002 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.
Other skills, from other repositories
frame-macos-notification
拟真 macOS 通知 banner + app icon + 标题正文, 适合 video overlay / 产品发布预告.
prototype-web
可点击的功能性 Web 原型, 含导航、英雄区、特性区、CTA.
waitlist-page
【模板: 等候名单页 / Waitlist】 【意图】为新产品 / 早鸟内测做一张极简等候页。 【布局】.
deck-course-module
暖纸背景 + Playfair, 左侧学习目标常驻, 含 MCQ 自测页.
deck-pitch
10 页融资 deck, 白底 + 蓝紫渐变 hero, traction 柱状, $X.XM ask.
conversational-ux
Design voice and conversational interfaces — dialog flows, error recovery, and persona. Use when the interface speaks and listens rather than being tapped. For graphical input collection, use form-design.