

Also from Kynth Studios


Also from Kynth Studios


Also from Kynth Studios
1# Repository Agent Guidance23Durable rules only. Perishable lane state — branch, milestone, known flakes,4closed investigations — lives in the private `codewhale-ops` repo, not here.5Read it there; don't trust memory of it.67## Intent is the artifact89Writing the code again is cheaper than recovering the code we wrote. Act on10that.1112- **Rewriting any part of this project is always in scope**, up to the whole13 thing. Nothing is load-bearing by virtue of existing. Argue a rewrite on14 merit, not sunk cost.15- **Use git; do not be governed by it.** A branch 600 commits behind is a note16 describing something we once wanted, not a debt. Conflict count is a signal to17 rewrite, not a task list.18- **A stranded lane becomes an issue, not a merge.** State the intent, the19 behavior wanted, and evidence worth keeping; reference the dead branch for20 provenance; abandon the branch; rebuild from current `main`.21- **Verify before you rebuild.** Grep for the symbols and behavior — not the22 commit — to check whether `main` already does it. Re-landing landed work is23 the failure mode this ethos creates, and it is the one you own.2425Limits: `main` stays protected and releases reproducible (never rewrite26published history, retag a shipped release, or force-push a shared ref);27contributor credit carries onto the rewrite; the do-not-delete guardrail below28still binds; and don't rewrite to avoid understanding.2930The four bullets above are the authoritative statement of this rule. Don't31restate them elsewhere — link here. (`docs/AGENT_ETHOS.md` is about stewardship32and workflow, not about this; it is not a longer form of this section.)3334## Build and test3536Always before pushing: `cargo fmt`, then targeted tests for the area.3738```sh39cargo test -p codewhale-config40cargo test -p codewhale-protocol41cargo test --workspace # full gate42cargo build --release -p codewhale-cli -p codewhale-tui # release build43```4445Crate-specific commands live in that crate's `AGENTS.md`. Environment quirks46(Cursor Cloud, keyless providers, dispatcher siblings) live in47`docs/ENVIRONMENTS.md`.4849Default branch is `main`. Committing directly to `main` is fine for release-lane50work — one reviewable concern per commit, with a real body. A fresh `codex/...`51branch or worktree is still right for an isolated or risky change.5253Commit as **WIP** unless you actually verified the behavior — built the binary,54ran the test, reproduced the fix. "Fixed" without evidence is worse than an55honest WIP.5657## Do-not-delete guardrail5859These are actively imported and have been repeatedly misflagged as dead code;60deleting them broke the build. Verify consumers with `rg` before believing any61dead-code audit:6263`tui/src/context_budget.rs`, `tui/src/model_registry.rs`,64`tui/src/prompt_zones.rs`, `tui/src/tools/remember.rs`, and the entire65`config/src/route/` directory.6667(`tui/src/memory.rs` was deliberately deleted in v0.9.4 — the native memory68store in `tui/src/native_memory.rs` is the surviving system; `tools/remember.rs`69is its capture path and stays.)7071## Surfaces that exist today7273Build only on these — removed machinery stays gone. The model-facing sub-agent74surface is **`agent` only**: the `agent_open`/`agent_eval`/`agent_close`/75`delegate_to_agent` variants, capacity/coherence/runtime-tag systems, lifecycle76tools, and runtime prompt/tag injection were all removed. The constitution77(`BASE_PROMPT` in `tui/src/prompts/text.rs`) is the sole base prompt.78Configurable sub-agent depth stays; add a new limit only when clearly needed,79and explain why.8081## Stewardship8283CodeWhale started as a DeepSeek-only harness; it is now about building the best84possible coding harness with an open-source community. Keep CodeWhale branding85and every model/provider first-class — none privileged.8687- Community PRs, issues, repros, logs, and reviews are maintainer evidence, not88 queue noise. Review from code, tests, linked issues, comments, and checks.89- **Credit is CI-enforced.** `Co-authored-by` trailers are for human90 contributors only — `scripts/check-coauthor-trailers.py` rejects bot/tool ones91 (Claude, codex, cursor, `noreply@anthropic.com`). Use canonical identities92 from `.github/AUTHOR_MAP`; note agent assistance in a plain commit body.93- Keep gates warm and dry-run unless Hunter explicitly approves enforcement.94- Leave unrelated edits by other people or agents intact.9596Full ethos: `docs/AGENT_ETHOS.md`. Issue-triage standard, release queue, and97harvest procedure live in the private `codewhale-ops` repo — they are98maintainer process, not contributor-facing contract.99
One repository carrying more than one format is the comparison this product exists for: does anyone actually write different content in each file, or is one a copy of the other?
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| Hmbown/CodeWhaleCLAUDE.md · 41k | CLAUDE.md | agent-behaviour | 16/100 | 12 days ago | |
| Hmbown/CodeWhalecrates/tui/AGENTS.md · 41k | AGENTS.md | testlint-formatdo-notagent-behaviour | 84/100 | today | |
| Hmbown/CodeWhalecrates/tui/locales/AGENTS.md · 41k | AGENTS.md | styleagent-behaviour | 43/100 | 12 days ago | |
| Hmbown/CodeWhaleweb/AGENTS.md · 41k | AGENTS.md | buildtestdo-notagent-behaviour | 66/100 | 12 days ago |
Same format, overlapping stack, ranked by quality.
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| TryGhost/Ghoste2e/AGENTS.md · 55k | AGENTS.md | setupteststylearch+2 | 100/100 | today | |
| mui/material-uiAGENTS.md · 99k | AGENTS.md | setupbuildtestlint-format+9 | 100/100 | 14 days ago | |
| code-yeongyu/oh-my-openagentpackages/web/AGENTS.md · 68k | AGENTS.md | setupbuildtestlint-format+6 | 100/100 | 13 days ago | |
| aaif-goose/gooseAGENTS.md · 53k | AGENTS.md | setupbuildtestlint-format+7 | 100/100 | 8 days ago | |
| deepseek-ai/deepseek-harnessnative/landlock-run/AGENTS.md · 104k | AGENTS.md | setupteststylearch+3 | 100/100 | today | |
| duckduckgo/content-scope-scriptsspecial-pages/AGENTS.md · 70 | AGENTS.md | buildteststylearch+3 | 100/100 | 14 days ago | |
| n8n-io/n8npackages/@n8n/agents/AGENTS.md · 201k | AGENTS.md | buildteststylearch+3 | 100/100 | 14 days ago | |
| elastic/elasticsearchx-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/ml/AGENTS.md · 78k | AGENTS.md | buildtestlint-formatstyle+2 | 100/100 | 14 days ago |
A badge carrying the measured quality of the strongest agent config file in this repository, out of 100. It reads from this index every time somebody loads your page, so it changes when the measurement changes and there is nothing to keep up to date. Free, no account, and the value is not something you or we can set by hand.
[](https://rulestack.kynth.studio/configs/hmbown-codewhale-agents)Would rather not hotlink us? Every badge is also served in shields.io’s endpoint schema, so shields renders the image and your readers never talk to our domain:
Published by Toolproof, the masthead over this index and eight others. The method behind the number is at toolproof.kynth.studio/methodology, and the whole thing is readable as JSON with no key at /api.