Thore CimbalandClaude Fable 5
e36ed337a7
Tracer bullet of the migration design (Gate 4, slice 1): pinned v0.1.1 baseline under docs/sources/upstream/ with provenance note, the Karpathy block moved verbatim to docs/sources/regelwerk/ (standing rule mapped onto the sources read-only mechanism), AGENTS.md assembled from the byte-true upstream sections plus the project section 6 (group rules condensed from the old CLAUDE.md), CLAUDE.md reduced to the upstream pointer, WORKFLOW.md and all four templates copied, schema.yaml extended (issue milestone/priority/status columns, component type, wiki area vision - all flagged in the header), validate.py and gen_status.py forked with marked extensions, pruefe_upstream_drift.py added, STATUS.md generated, CI gains the offline validate job, README directory link defused. Verified: validate 0 errors 0 warnings (the three pre-existing directory-link errors are gone), gen_status --check current, drift check 0 findings, baseline byte-identical to the reference checkout (10/10 files), four negative tests fire (WIP limit 3x in-progress, waiting without wartegrund, component slug mismatch, single-byte drift in WORKFLOW.md). gen_status needs Python >= 3.10 locally (write_text newline) - noted for the design AAR. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2.5 KiB
name, description, license
| name | description | license |
|---|---|---|
| karpathy-guidelines | Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria. | MIT |
Karpathy Guidelines
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Tradeoff: These guidelines bias toward caution over speed. For trivial tasks, use judgment.
1. Think Before Coding
Don't assume. Don't hide confusion. Surface tradeoffs.
Before implementing:
- State your assumptions explicitly. If uncertain, ask.
- If multiple interpretations exist, present them - don't pick silently.
- If a simpler approach exists, say so. Push back when warranted.
- If something is unclear, stop. Name what's confusing. Ask.
2. Simplicity First
Minimum code that solves the problem. Nothing speculative.
- No features beyond what was asked.
- No abstractions for single-use code.
- No "flexibility" or "configurability" that wasn't requested.
- No error handling for impossible scenarios.
- If you write 200 lines and it could be 50, rewrite it.
Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.
3. Surgical Changes
Touch only what you must. Clean up only your own mess.
When editing existing code:
- Don't "improve" adjacent code, comments, or formatting.
- Don't refactor things that aren't broken.
- Match existing style, even if you'd do it differently.
- If you notice unrelated dead code, mention it - don't delete it.
When your changes create orphans:
- Remove imports/variables/functions that YOUR changes made unused.
- Don't remove pre-existing dead code unless asked.
The test: Every changed line should trace directly to the user's request.
4. Goal-Driven Execution
Define success criteria. Loop until verified.
Transform tasks into verifiable goals:
- "Add validation" → "Write tests for invalid inputs, then make them pass"
- "Fix the bug" → "Write a test that reproduces it, then make it pass"
- "Refactor X" → "Ensure tests pass before and after"
For multi-step tasks, state a brief plan:
1. [Step] → verify: [check]
2. [Step] → verify: [check]
3. [Step] → verify: [check]
Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.