Stats
Actions
Tags
From forge
Use when about to edit, create, or delete files as part of an implementation task — before writing any code. Rates each file's risk level and adjusts the task size accordingly.
How this skill is triggered — by the user, by Claude, or both
Slash command
/forge:risk-classifyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Rate every file the task will touch before writing any code. Output the classification, then proceed.
Rate every file the task will touch before writing any code. Output the classification, then proceed.
| Rating | Criteria |
|---|---|
| 🔴 | Filename or content contains: auth, jwt, token, crypto, hash, password, payment, charge, invoice, migration, schema, delete, drop, public API, webhook, signature, secret |
| 🟡 | Existing business logic files, service files, repository files, controller files, state management, SQL queries |
| 🟢 | Test files, documentation, config files, new files being created from scratch, CSS/style files |
Any 🔴 file escalates the task to Large and triggers 3 adversarial reviewers.
📁 Risk classification:
🔴 src/webhooks/payments.ts — payment processing
🟡 src/services/order.service.ts — business logic
🟢 src/webhooks/payments.test.ts — test file
Task size escalated to Large (🔴 file present).
Adversarial review: 3 reviewers.
Output this block before touching any files.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub addeelnayyer/forge --plugin forge