FableClaudeMDForOpus

KO | EN

gitlite — search

#python #llm #golang #rust #linux #macos #awesome #kubernetes #awesome-list #security #ios #mcp #openai #open-source

FableClaudeMDForOpus

(★ 246)

A Fable written Claude MD for Opus/Sonnet to give it some of the quality methodology that Fable uses

파일 탐색기

최종 버전 다운로드 (.zip)

- - _FORMAT.md
  - CODE.md
  - DEBUG.md
  - EFFICIENCY.md
  - PLAN.md
  - SESSION.md
  - TRAPS.md
  - VERIFY.md
- research-digest.md
- review-digest.md
- STATE.md
CLAUDE.md
MIGRATE.md
README.md

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Guardrails Kit v1.0

A portable CLAUDE.md + documentation set that makes Claude Opus / Sonnet operate as close to
frontier (Fable) level as possible inside Claude Code: fewer logic errors, fewer introduced bugs,
fewer wasted tokens. It works by converting the implicit judgment a stronger model applies
automatically into explicit, checkable, event-triggered procedures a weaker model can execute
mechanically.

## What's in the kit

| File | Role |
|---|---|
| `CLAUDE.md` | Always-loaded core: 12 iron rules, an event-phrased routing table, 4 hard stops. Deliberately small (~45 kit lines, hard cap 60) — always-on compliance is roughly constant-sum, so every extra line taxes obedience to all the others. |
| `docs/guardrails/PLAN.md` | Before starting non-trivial work: TASK block, premise check, prior-art search, baseline, decomposition, ask-vs-decide. |
| `docs/guardrails/CODE.md` | While editing: read-before-edit gates, twin/generated-file checks, and the REFERENCE SWEEP procedure (RS1-RS5). |
| `docs/guardrails/TRAPS.md` | Lookup tables for the classic reasoning traps: dates, epochs, mutation-vs-copy, async, floats/money, sort, division/modulo, regex, familiar-API lookalikes, closures, boolean logic. Read on demand via CODE.md C7. |
| `docs/guardrails/DEBUG.md` | When anything fails: reproduce-first loop, CAUSE line, failed-attempts ledger with the ESCALATION LADDER, red-flag table keyed on the model's own rationalization phrases. |
| `docs/guardrails/VERIFY.md` | Before claiming done/committing: 12-item echo protocol — every claim needs output quoted from a real tool result in the same turn. |
| `docs/guardrails/EFFICIENCY.md` | Token/context discipline as paired rules: every "read less" rule has a "read enough" floor. |
| `docs/guardrails/SESSION.md` | Long-session survival: docs/STATE.md template (S2), same-turn update triggers (S3), post-compaction recovery (S1), ANCHOR/DETOUR/DECISION ledger keywords. |
| `docs/guardrails/_FORMAT.md` | Authoring contracts for editing the kit itself (budgets, trigger phrasing, single-sourcing, sanctioned iron-rule pairs). |
| `MIGRATE.md` | The transport procedure for retrofitting a project that already has a CLAUDE.md — line-accounted, backup-first, verbatim-carry, user-checkpointed, idempotent, with an UPGRADE mode. |

Not part of the installable kit (kit-source materials only): `docs/research-digest.md`
(the 155-finding failure-mode research behind every rule), `docs/review-digest.md`
(the 193-finding adversarial review that hardened it), and `docs/superpowers/specs/`
(the design record).

## Install — fresh project (no existing CLAUDE.md)

From the project root:

```powershell
Copy-Item <kit>/CLAUDE.md CLAUDE.md
New-Item -ItemType Directory -Force docs/guardrails | Out-Null
Copy-Item <kit>/docs/guardrails/*.md docs/guardrails/
```

Then fill the `## Project` section of CLAUDE.md with your run/test commands and hard project
constraints (cap: 40 lines — everything conditional goes in `docs/guardrails/PROJECT.md` with a
pointer line in `## Project`). Never edit inside the `BEGIN/END KIT` markers.

## Install — project with an existing CLAUDE.md

Tell the model (Opus is fine — the procedure is designed for it):

> Read MIGRATE.md in <kit path> and execute it exactly, phase by phase.

MIGRATE.md is built so nothing is lost: snapshot first, every original line gets a logged
disposition, kit files are installed by per-file copy (never retyped), rule conflicts are surfaced
instead of silently resolved, and it stops for your approval before installing kit docs or
composing the new CLAUDE.md. Re-running it on a migrated project is detected and switches to
UPGRADE mode.

## Design principles (why it looks like this)

1. **Lean core + on-demand playbooks.** Only ~45 kit lines are always loaded. The extensive
   material is read at the moment its trigger fires — that is how "extensive" and "fewer tokens"
   coexist.
2. **Event-phrased routing.** Models route on what they literally experience ("a test failed",
   "about to type done"), not on topic labels ("debugging", "verification"). A doc that never
   gets opened is worth zero, so the routing table is the most engineered block in the kit —
   and its contract makes the doc-Read the next tool call, alone in its message.
3. **Paste, don't check.** Every rule produces a transcript artifact (a pasted grep, a quoted
   summary line, a `TRIGGER:`/`ANCHOR:`/`V3: PASS` line), and VERIFY requires quoted lines to
   exist verbatim in a tool result in the same turn. Compliance is visible; "ensure" is not.
4. **Numbers, not judgment.** "10 messages", "2 failures", ">50 hits", ">300 lines" — weaker
   models comply with countable thresholds and rationalize their way around graded ones.
5. **Prohibitions carry replacements.** Every NEVER has its replacement on the same line,
   because a banned action with no named alternative gets taken anyway under pressure.
6. **Single source with sanctioned compression.** Each rule lives in exactly one file; a
   CLAUDE.md iron rule may compress a doc rule, but shared trigger lists stay byte-identical
   (_FORMAT.md F7 lists the sanctioned pairs).

## Auditing compliance

The kit's markers are greppable in any session transcript. To see which rules fired and which
were skipped, search a transcript for: `TRIGGER:`, `GOAL:`, `FILES:`, `EST:`, `DONE-WHEN:`,
`BASELINE:`, `ASSUMPTION:`, `PLAN CHANGE:`, `CAUSE:`, `WORKAROUND:`, `ATTEMPT `, `ANCHOR:`,
`DETOUR(`, `RETURNING:`, `DECISION:`, `CONSTRAINT CHECK:`, `HANDLED FAILURES:`,
`NOTED (not done)`, `EDITED-UNVERIFIED`, `CANNOT-REPRODUCE`, `SIGNATURE UNVERIFIED`,
`P1:`–`P9:`, `C1:`–`C15:`, `D1:`–`D10:`, `V1:`–`V12:`, `E1:`–`E17:`, `S1:`–`S7:`, `RS1`–`RS5`.
Missing markers at the moments their triggers occurred are the non-compliance you should tune for.

## Upgrading the kit

Kit text in an installed project lives inside `` and
`` markers and as verbatim files under `docs/guardrails/`.
Upgrades are wholesale block/file swaps — see UPGRADE mode (U0–U4) at the bottom of MIGRATE.md —
which is exactly why project content must never be interleaved into kit blocks, and why kit
files must never be paraphrased. When editing kit content itself, follow
`docs/guardrails/_FORMAT.md`.

## Upgrade notes

- v1.0 — initial release: 8 guardrail docs + CLAUDE.md core + MIGRATE.md, hardened by a
  13-reviewer adversarial pass (193 findings applied: observable routing events, single-source
  ownership with byte-identical trigger lists, canonical status vocabulary, cross-platform
  command pairs, TRAPS.md split, MIGRATE collision/idempotency/verbatim repairs).

FableClaudeMDForOpus

파일 탐색기

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

FableClaudeMDForOpus

파일 탐색기

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README