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 rules/zxpmail/reqforge/reqforge-coregit clone --depth 1 https://github.com/zxpmail/ReqForgeWhat 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 | $0.00714 | $0.00714 |
| Opus 5 | $0.00357 | $0.00357 |
| Sonnet 5 | $0.00143 | $0.00143 |
| Haiku 4.5 | $0.00071 | $0.00071 |
Grade A, and why
reqforge-core 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forge Core Development Rules
TDD First (RED-GREEN-REFACTOR)
Tests must be written before functional code. No code without tests is allowed to be committed.
- RED: First write a test that will fail
- GREEN: Write minimal code to make the test pass
- REFACTOR: Refactor while keeping the test green
Code Standards
- Single file does not exceed 300 lines — split by responsibility
- TypeScript strict mode, no
any(useunknown+ type guards) - Naming: Components PascalCase, functions/variables camelCase, files kebab-case, constants UPPER_SNAKE_CASE
- Each file has a single responsibility with a clear external interface
- React prefers function components + Hooks, no class components
- Styles prefer Tailwind, don't write custom CSS unless necessary
- No unrelated refactoring — only touch what needs changing
- YAGNI: don't write code for hypothetical future requirements
- Follow the existing codebase style — don't force your own preferences
Verification Is Evidence (Hard Gate)
Every completion declaration must include the verification command and its output executed in the same message. "It's done" with no verification command is an invalid declaration. No on-the-spot verification, no completion.
Four-Step Phase Verification
After all Tasks in a Phase are complete:
- Code Review — cross-reference delivery checklist item by item
- Test Completeness — all features implemented, no omissions
- Compile Check —
tsc --noEmitzero errors - Functional Test — start dev server, verify no errors, test both new and existing features
If any step fails → fix → restart from Step 1. One pass is rarely enough.
Memory System
memory/project-memory.md— Architecture facts, constraints (permanent)memory/decisions-log.md— ADR-format decision records (permanent)memory/task-history.md— Recent task summaries, max 30 entries (rolling)
After every completed Task: append to task-history.md. If a technical decision was made, append to decisions-log.md. If architecture facts changed, update project-memory.md.
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 · 62 lines · 714 tokens per session scan A 66a932072fee
reqforge-core is a cursor rule published in the GitHub repository zxpmail/ReqForge (18 stars, last pushed 11d ago), licensed MIT. It adds 714 tokens to every session, about $0.0036 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 2026-08-30.
Other cursor rules, from other repositories
language-agnostic-patterns
Language-agnostic programming patterns: SOLID, design patterns, clean code, and architecture. Load when refactoring, designing abstractions, or reviewing structure — not for everyday syntax.
cursor-tools-mastery
Cursor 3.7 runtime guide: choose the right tool, canvases, Design Mode, /worktree, /best-of-n, Await, and parallel execution where safe.
cursor-agent-orchestration
Cursor 3.7 orchestration guide: when to plan, when to delegate, nested subagents, multi-environment handoffs, /best-of-n, and Await for long-running branches.
fable5-reasoning
Fable 5 reasoning protocols: task interpretation, risk-first decomposition, approach selection, interleaved thinking, hypothesis ledgers, premortems, calibration, and the stuck-strategy ladder. Load for complex, ambiguous, or long-horizon tasks, for debugging strategy, or whenever progress stalls.
cursor-mcp-optimization
Cursor 3.7 MCP optimization: browser Design Mode, canvases, Figma, Cloudflare tools, MCP Apps structured content, and direct action patterns.
minimax-m3-self-evolution
MiniMax M3 self-evolution harness: iterative refinement loops, compress-before-iterate, autonomous debugging, and recursive improvement patterns.