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-apinpx skills add ForgeyClap/claude-forge --skill forge-apigit clone --depth 1 https://github.com/ForgeyClap/claude-forgeWhat 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 | $0.00047 | $0.01467 |
| Opus 5 | $0.00023 | $0.00733 |
| Sonnet 5 | $0.00009 | $0.00293 |
| Haiku 4.5 | $0.00005 | $0.00147 |
Grade A, and why
forge-api 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 2d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forge playbook — Contract-first API
Do not duplicate ECC skills — defer to: forge-integration (when the API mainly wires external services / webhooks), forge-payments (any money/charge/gateway-webhook endpoint), forge-fullstack (when the API is one tier of a full app). This file is orchestration only.
This is an auth + data-boundary domain. The contract (OpenAPI / JSON Schema / GraphQL SDL / gRPC proto) is the source of truth and comes first; the implementation is validated against it. architect owns the contract, integration-boss / build-boss implement, security-boss reviews auth/authz (a genuine high-risk trigger), test-boss owns the contract tests. Never claim "auth enforced" or "contract-compliant" without a negative test that proves it.
Hard rules (non-negotiable)
- Contract first, single source of truth. Define the API contract (OpenAPI 3.x, JSON Schema, GraphQL SDL, or a gRPC
.proto) BEFORE implementing handlers. Commit it; the implementation is generated from or validated against it. Drift between contract and behavior is a bug, not a detail. - Input validation at every boundary. Validate every request — body, path params, query, headers — against the schema. Reject invalid input with a
4xxand a consistent error envelope; enforce type, size, and range limits. Never trust client input; never pass it unvalidated into a query, filesystem, or shell. - Auth on every protected route, server-side, deny-by-default. Authentication AND authorization are enforced in the service on each protected route — not only at a gateway, not "assumed from the frontend". New routes default to protected; opening one is a deliberate, reviewed act. Per-route scope/role checks; a request without valid credentials gets
401/403, proven by a negative test. - Explicit versioning. The API is versioned (path
/v1, header, or media type). No breaking change ships inside an already-published version; breaking changes get a new version + a stated deprecation path. - Idempotency on unsafe retryable writes.
GET/PUT/DELETEare idempotent by design; retryablePOSTcreates (orders, payments, side-effecting actions) accept a clientIdempotency-Keyso a retry or double-submit does not duplicate the effect. - Rate limits + quotas. Per-client / per-key rate limiting with
429+Retry-After; abuse and runaway clients are bounded. Consistent error envelope; pagination on list endpoints (no unbounded result sets); secrets in env — never in responses, error bodies, or logs.
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.
- 2d ago First seen · 45 lines · 47 tokens per session scan A f4a36a6ee4cc
forge-api is a skill published in the GitHub repository ForgeyClap/claude-forge (2 stars, last pushed 29d ago), licensed MIT. It adds 47 tokens to every session and 1,467 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
zh-output-example
Reviews Chinese skill documentation for publication readiness. Invoke when the user asks in Chinese to review a skill.
argent-tv-interact
Control and inspect TV apps via argent — Apple TV (tvOS), Android TV (leanback), and Amazon Fire TV (Vega). Boot the target, read focus, navigate with the D-pad remote, type, screenshot, and on Vega debug the JS runtime (evaluate, console logs, network inspector). Use when a task targets a TV (runtimeKind "tv", or…
review-offered-task
Review a task that has been offered to you and decide whether to accept or reject it.
company-hiring-intelligence
Reverse-engineer what a company is building by scraping their job postings, careers page, LinkedIn Jobs, and engineering blog using TinyFish web agents. Use whenever a user wants to understand a company's strategic direction from hiring signals, do competitive intelligence, figure out a tech stack from job…
文档协作
引导用户通过结构化的文档共同编写工作流程。当用户想撰写文档、提案、技术规范、决策文档或类似结构化内容时使用。该工作流程帮助用户高效传递上下文，通过迭代优化内容，并验证文档对读者有效。当用户提到写文档、创建提案、起草规范或类似文档任务时触发。.
aidd-dev:08:for-sure
Iterative agent loop that tracks attempts and retries until a success condition is met. Use when the user says "for sure", "make sure", "keep trying until", "loop until done", "don't stop until", or needs guaranteed completion of a task with explicit success criteria.