HERO — Anti-OverDefense 🧱

· · · ·

🧱 You asked for a feature. It built a fortress around the feature, and never got to the feature.

你让它写个功能,它给你造了座堡垒,然后功能一直没写完。

The shapes look like an agent optimising for not being blamed rather than for the work being good — a hypothesis that fits what we observed, not an established account of how these models are trained. Either way the shapes are real, there are four of them, and their initials are the name: Hashing, Edge cases, Rubrics, Overbuild. Naming them makes it possible to say which one just happened instead of arguing about vibes.

This repository is a short block you paste into your agent's config, plus a catalogue of the behaviours it is meant to stop. It works with Claude Code, Codex, Antigravity, Cursor, GitHub Copilot, Windsurf and Gemini CLI — anything that loads a config file without being asked. There is nothing to install for those rules. For long goals in Codex or Claude Code, the optional Goal Pulse skill enables execution reminders by default when you start or resume a Goal after installing it. For a fresh conversation, Session Handoff preserves the task in project files and gives the next session a concrete place to continue.

🧬 Generalised out of ARIS (~13.7k★), whose cross-model reviewer is genuinely good — and also spent a measurable share of its output proposing hashes nobody reads. HERO keeps the value and drops the tax.

🧭 What's inside

Three complementary paths: HERO Rules keep work focused, Goal Pulse keeps an active goal moving, and Session Handoff carries useful state into a fresh conversation. The last is a file-based workflow: prepare the handoff, open a new session, and continue from the recorded evidence and next action. The diagram shows the Codex Python timer; Claude Code uses native /loop for the same Goal Pulse role.

📢 What's changed

Only things that change what you should do. Not rewording, not layout — a changelog of everything buries the one line that mattered.

• 2026-11-02 — Session Handoff — the third HERO path. The Markdown skill for Codex and Claude Code saves the objective, evidence, decisions, active work, and next action, then guides a fresh session to continue from those files. It reuses existing recovery notes and needs no Python. Together with HERO Rules and Goal Pulse: focus, momentum, continuity.

• 2026-11-01 — Goal Pulse for Codex. Once installed, the skill and local timer activates by default when you start or resume a Goal, reminding the main agent to collect finished work and tackle recurring blockers. Bilingual reminders run every 30 minutes; paused Goals stay silent. The core HERO block is unchanged.

  Full Chinese reminder sent by the timer:

  中文：以最快的有效速度持续推进当前 Goal，不允许磨洋工、偷懒、空转，或拿计划和汇报代替执行。

  worker/子代理一交付完整产物，立即读取、核验，并推进到下一项已授权的验收或交接步骤，不得搁置或等整批。

  同一卡点连续两轮无实质进展，主代理立即接管具体缺口或重新拆分，同时推进其它独立任务。

  复用已有成果与有效验证，砍掉重复检查、无收益的 corner case 和包装。

  不得降低原目标、难度和验收标准，不得越过授权和预算；直接行动，无需专门回复本提醒。

• 2026-09-03 — 📣 Rule 7 gains the press-release principle — re-paste it. A paper is a launch, not a progress report: lead with the strongest result; an unfavorable number that is a tradeoff is explained as one, one that is not is stated plainly, and either way it stays in the table — never narrated as a defeat. Adopted from anti-defensive-writing-Skill by @Adkid-Zephyr, which works the principle out in full for papers.
• 2026-09-02 — 🏃 The block changed — re-paste it. Rule 8: momentum is part of the deliverable — when several options are reasonable, pick one and note the tradeoff instead of handing back a menu. Rule 7 now also bans process traces: intermediate errors, abandoned approaches and AI-assistance markers are not content; rule 9 spells that out in full. Both distilled from a working ML researcher's config — the original is in examples/. New sibling named: stalling.
• 2026-08-26 — ✍️ Rule 7, plus two new siblings: defensive prose. Caveats scattered through every paragraph until the work reads as an apology (SIB-003); instructions confessed into the product — told "do not mention X", it writes "we do not address X" (SIB-004). Rule 7: deliverable text is not a defense transcript. Distilled from repairing a writing pipeline whose papers had turned into confessions.
• 2026-08-12 — ⏳ The block fades on long runs. Three causes, different answers: something more specific is contradicting it (repeating won't help), it is still there but buried under hours of output (repeating helps), or compaction thinned it (re-emitting helps). Don't set an hourly timer — context grows with work, not with minutes. How to tell which · hook route.
• 2026-08-11 — 📓 Two cases from long-running quant research (#1, thanks @leisurexhx). HERO-R-006 — audit loops that ate the experiment they were auditing. HERO-O-005 — a permanent version tree for every recoverable failure. The best line is in neither: a general "only do the checks you need" lost to a twelve-stage workflow in the same prompt.
• 2026-08-11 — 🧱 The block changed — re-paste it. New rule 6: this never overrides security or migration work you actually asked for. Eight examples added, two of them marked ✓ meaning report this, don't dismiss it. There is also a short version at half the size.
• 2026-08-11 — 🧱 HERO published.

⚠️ Pasted the block before any of the above? Yours is stale. The install command won't fix it — the guard that stops it appending twice also stops it replacing. Delete your old === SCOPE LIMITS ... === section and run it again.

🚀 Quickstart

Paste this into the file your agent loads automatically — CLAUDE.md, AGENTS.md, .github/copilot-instructions.md, .cursorrules. See hosts/ for the full table.