菜单
Generate a new database-backed MCP tool environment from a scenario description. Use when user says "forge an environment", "generate tool environment", "create an MCP backend", "scaffold database tools", "build a sandbox", or "synthesize an environment from scratch".
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | forge |
| description | Generate a new database-backed MCP tool environment from a scenario description. Use when user says "forge an environment", "generate tool environment", "create an MCP backend", "scaffold database tools", "build a sandbox", or "synthesize an environment from scratch". |
| metadata | {"author":"Fred Bliss","version":"0.3.1","last_verified":"2026-04-02"} |
Interactive environment generation. Walks through the full synthesis pipeline: scenario description -> tasks -> schema -> seed data -> API spec -> server code -> verifiers. Each step produces a file in the output directory.
/env-forge:forge <description of the platform/domain>
/env-forge:forge
Examples:
/env-forge:forge a volunteer matching platform where organizations post opportunities and volunteers sign up based on skills and availability/env-forge:forge pet adoption agency with animal profiles, applications, foster homes, and veterinary records/env-forge:forge then describe the domain interactivelyIf the user provides a description, refine it. If not, ask:
Write the scenario to .env-forge/environments/<name>/scenario.json:
{
"name": "volunteer_match",
"description": "A volunteer matching platform where...",
"tasks": [],
"created_at": "2026-02-23T12:00:00Z"
}
Before generating from scratch, search the AWM-1K catalog for a structurally similar domain:
uv run python env-forge/scripts/catalog.py --search "<domain keywords>"
If a close match exists, review its structure with --details:
Use the closest match as a structural reference -- adapt its patterns to the new domain, don't copy verbatim. If no match is close enough, proceed from scratch.
Generate 10 realistic, diverse tasks following the task-first methodology from the env-forge skill. Each task is a single sentence with all parameters included.
Coverage requirements:
Present tasks to user for review. Allow additions, removals, or modifications. Update scenario.json with finalized task list.
Design SQLite tables to support ALL tasks. Follow patterns in references/schema_patterns.md:
Write DDL to .env-forge/environments/<name>/schema.sql.
Present table list and relationships to user for review.
Generate INSERT statements that make every task executable. Follow coverage rules in references/schema_patterns.md:
Write to .env-forge/environments/<name>/seed_data.sql.
Design RESTful endpoints for all tasks. Follow references/api_design_rules.md:
Write to .env-forge/environments/<name>/api_spec.json.
Present endpoint list to user for review.
Generate complete, executable FastAPI application. Follow references/fastapi_mcp_template.md:
Write to .env-forge/environments/<name>/server.py.
Generate verification functions for each task. Follow references/verification_patterns.md:
Write to .env-forge/environments/<name>/verifiers.py.
Create and seed the SQLite databases:
mkdir -p .env-forge/environments/<name>/db
sqlite3 .env-forge/environments/<name>/db/current.db < .env-forge/environments/<name>/schema.sql
sqlite3 .env-forge/environments/<name>/db/current.db < .env-forge/environments/<name>/seed_data.sql
cp .env-forge/environments/<name>/db/current.db .env-forge/environments/<name>/db/initial.db
Write pyproject.toml for the environment:
[project]
name = "env-forge-<name>"
version = "0.1.0"
requires-python = ">=3.11"
dependencies = [
"fastapi>=0.115.0",
"uvicorn>=0.30.0",
"sqlalchemy>=2.0.0",
"fastapi-mcp>=0.3.0",
"pydantic>=2.0.0",
]
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"
Run structural validation:
uv run python env-forge/scripts/validate_env.py .env-forge/environments/<name>/
Display:
cd .env-forge/environments/<name> && uv run python server.pyverifiers.pyIf any generation step produces errors (syntax, import, schema mismatch):
Before starting, assess whether the domain produces useful synthetic environments. Flag if the domain's core value depends on non-synthesizable data (long-form content, media, AI inference, real-time feeds). See the suitability table in the env-forge skill.
Full environment directory at .env-forge/environments/<name>/ with all files listed above. The environment is self-contained and runnable.
Opt the current project into down-tier model delegation by installing a standalone .claude/rules/model-delegation.md rule, optionally with pre-shaped .claude/agents/ definitions (fast-executor, task-coder) and an optional agent-state feedback layer. The rule tells Claude to route well-specified data and coding tasks to a cheaper model in a subagent, keeping judgment-heavy work in the main loop. Use when the user says "set up model routing", "set up model delegation", "use cheaper models for subagents", "delegate to a lower-power model", or wants to remove or update that rule.
Pre-share scan for leaked secrets and privacy-sensitive content. Wraps simonw/scan-for-secrets for literal matching (with JSON/URL/HTML/backslash/unicode escape variants) and composes a ripgrep regex pass for shape-based leaks (other users' home paths, emails, IPv4, MAC, JWTs, API keys). Use when user says "scan for secrets", "check for leaked credentials", "pre-share scan", "redact home paths", "PII scan", "strip my username from transcripts", "scan logs before publishing", "check before sharing", "audit before commit", "find leaked API keys", "scan agent transcript".
Convert JavaScript and TypeScript projects to use bun instead of npm/yarn/pnpm. Use when user asks to "migrate from npm to bun", "convert to bun", "switch package managers", "replace npm", or "use bun instead". Includes conversion tables and lock file migration. Core conventions auto-loaded via SessionStart hook; invoke /dev-conventions:bun-tooling for full reference.
Audit project dependencies for known vulnerabilities using uv audit (Python) and bun audit (JS/TS). Use when user says "audit dependencies", "check for vulnerabilities", "security audit", "check deps", "are my packages safe", "CVE check", "dependency security", "vulnerability scan". Invoke with /dev-conventions:dep-audit.
Documentation conventions: last-updated dates, lowercase filenames, internal/ subfolder for non-shared docs, session logs, and documenting the "why". Use when creating or updating documentation, READMEs, or design docs. Invoke with /dev-conventions:doc-conventions. Triggers on "write docs", "update README", "document this", "add a design doc", "session log".
Detailed Python/uv conversion reference. Core conventions auto-loaded via SessionStart hook; invoke /dev-conventions:python-tooling for full conversion tables. Use when you need the complete uv command mapping, version pinning strategy, or lock file workflow.