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 skills add cacheplane/dawnai --skill cligit clone --depth 1 https://github.com/cacheplane/dawnaiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cacheplane/dawnai/cli)<a href="https://agentmods.dev/skills/cacheplane/dawnai/cli"><img src="https://agentmods.dev/badge/skills/cacheplane/dawnai/cli/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cacheplane/dawnai/cli"><img src="https://agentmods.dev/badge/skills/cacheplane/dawnai/cli.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00048 | $0.00203 |
| Opus 5.5 | $0.00019 | $0.00081 |
| Sonnet 5.5 | $0.00010 | $0.00041 |
| Haiku 4.5 | $0.00005 | $0.00020 |
Grade A, and why
dawn 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 1mo 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.
What it actually says
Dawn
Dawn is the TypeScript meta-framework for LangGraph. Agents and workflows are
file-system routes under src/app/, with co-located tools, generated types, and
durable threads.
Source of truth
The complete, version-matched Dawn documentation ships inside this package at
docs/. Always read the bundled docs — they match the installed version exactly.
- Start with
docs/README.md(the index and recommended reading order). - Or run
dawn docsto list topics anddawn docs <topic>to read one (for example,dawn docs tools).
Do not rely on this file's prose for API detail; read the bundled docs first.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- CHANGELOG.md 104 KB
- package.json 2.6 KB
- README.md 3.3 KB
- scripts/generate-docs.mjs 2.1 KB runs code
- src/commands/add.ts 4.6 KB runs code
- src/commands/build.ts 3.1 KB runs code
- src/commands/check.ts 8.7 KB runs code
- src/commands/dev.ts 1.3 KB runs code
- src/commands/docs.ts 3.2 KB runs code
- src/commands/eval.ts 8.6 KB runs code
- src/commands/inspect.ts 6.8 KB runs code
- src/commands/memory.ts 14 KB runs code
- src/commands/routes.ts 1.3 KB runs code
- src/commands/run.ts 3.0 KB runs code
- src/commands/start.ts 2.5 KB runs code
- src/commands/test.ts 5.6 KB runs code
- src/commands/typegen.ts 1.3 KB runs code
- src/commands/verify.ts 12 KB runs code
- src/fetch-exports.ts 2.8 KB runs code
- src/index.ts 3.9 KB runs code
- src/lib/build/deployment-config.ts 965 B runs code
- src/lib/build/targets/edge-capabilities.ts 15 KB runs code
- src/lib/build/targets/edge-modules-emitter.ts 6.2 KB runs code
- src/lib/build/targets/hono.ts 5.3 KB runs code
- src/lib/build/targets/index.ts 2.2 KB runs code
- src/lib/build/targets/langsmith.ts 3.4 KB runs code
- src/lib/build/targets/modules-emitter.ts 19 KB runs code
- src/lib/build/targets/node.ts 7.4 KB runs code
- src/lib/build/targets/thread-access-probe.ts 1.6 KB runs code
- src/lib/build/targets/vercel-config.ts 7.9 KB runs code
- src/lib/build/targets/vercel-node-compat.ts 3.0 KB runs code
- src/lib/build/targets/vercel-output.ts 11 KB runs code
- src/lib/build/targets/vercel.ts 4.3 KB runs code
- src/lib/build/targets/web-runtime.ts 34 KB runs code
- src/lib/dev/abortable-iterable.ts 2.1 KB runs code
- src/lib/dev/agui-handler.ts 22 KB runs code
- src/lib/dev/allocate-port.ts 740 B runs code
- src/lib/dev/boot-runtime.ts 1.1 KB runs code
- src/lib/dev/classify-change.ts 2.7 KB runs code
- src/lib/dev/cors.ts 7.9 KB runs code
- src/lib/dev/dev-child.ts 8.4 KB runs code
- src/lib/dev/dev-session.ts 11 KB runs code
- src/lib/dev/health.ts 1.1 KB runs code
- src/lib/dev/load-app-env.ts 1.5 KB runs code
- src/lib/dev/load-env.ts 1.6 KB runs code
- src/lib/dev/memory-handler.ts 3.6 KB runs code
- src/lib/dev/middleware-node.ts 7.0 KB runs code
- src/lib/dev/middleware.ts 3.4 KB runs code
- src/lib/dev/node-web-adapter.ts 4.3 KB runs code
- src/lib/dev/parked-route.ts 7.9 KB runs code
- src/lib/dev/parse-port.ts 621 B runs code
- src/lib/dev/pending-interrupts.ts 8.0 KB runs code
- src/lib/dev/request-context.ts 893 B runs code
- src/lib/dev/resolve-env-path.ts 1.0 KB runs code
- src/lib/dev/run-registry.ts 5.6 KB runs code
- src/lib/dev/runtime-fetch-core.ts 125 KB runs code
- src/lib/dev/runtime-fetch-handler.ts 1.3 KB runs code
- src/lib/dev/runtime-registry-core.ts 4.0 KB runs code
- src/lib/dev/runtime-registry.ts 928 B runs code
- src/lib/dev/runtime-server.ts 11 KB runs code
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.
- 1mo ago First seen · 22 lines · 48 tokens per session scan A 12bbf18f7742
dawn is a skill published in the GitHub repository cacheplane/dawnai (327 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 203 once invoked, about $0.0002 per session on Opus 5.5. 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 skills, from other repositories
code-review
Perform a structured code review of changes, checking for correctness, style, tests, and potential issues.
credit-note-fixer
Fix the tiny credit-note formatting bug and rerun the exact targeted test command.
om-troubleshooter
Diagnose and fix standalone Open Mercato bugs across scope, commands, locking, fields, generated registries, UI hydration, cache/search, bootstraps, queues, and providers. Use for "fix bug", "why does this fail", "regression", "debug", "napraw błąd", or a failing test.
error-analysis
Evals-first error analysis for LLM apps: clusters real Langfuse or JSONL traces into a human-confirmed failure taxonomy with counts, then recommends binary pass/fail evals for recurring named modes. Use to learn what to measure before writing evals. Not for CI failures.
paired-probe
Refuse a verdict a probe did not earn. Runs a check where the fault IS present and where it is NOT, and blocks the answer when both arms print the same thing, because a check that cannot disagree with you has measured nothing. Also catches the zero-sample sweep that reads as "clean" and the swallowed error that reads…
systematic-debugging
A structured process for investigating and fixing software problems through four stages. It covers systematic debugging principles and rules.