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 agentmods add skills/auravixstudio/caelo/designnpx skills add AuraVixStudio/caelo --skill designgit clone --depth 1 https://github.com/AuraVixStudio/caeloWrote 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/auravixstudio/caelo/design)<a href="https://agentmods.dev/skills/auravixstudio/caelo/design"><img src="https://agentmods.dev/badge/skills/auravixstudio/caelo/design.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00022 | $0.00369 |
| Opus 5 | $0.00011 | $0.00185 |
| Sonnet 5 | $0.00004 | $0.00074 |
| Haiku 4.5 | $0.00002 | $0.00037 |
Grade A, and why
Design doc (write -> review loop) 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 6d 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
Design doc (write -> review loop)
Use this to produce a vetted design document for a non-trivial task before implementing.
You iterate a design-doc-writer (writes in an isolated copy) against a
design-doc-reviewer (read-only critique) until the reviewer approves.
Loop
- Draft.
delegateadesign-doc-writerwith the problem, constraints and any relevant code paths. It writes a Markdown design doc in its isolated copy. - Review.
delegateadesign-doc-reviewerto critique that document. The reviewer ends withVERDICT: APPROVEorVERDICT: REVISE. - Revise. If the verdict is REVISE,
delegateadesign-doc-writeragain with the reviewer's concerns and instruct it to address every point. - Repeat steps 2–3 until the verdict is APPROVE, or you reach the round limit (default 3; honor an "effort N" of 1–5 as N rounds).
- Summarize. Report the agreed design (path + key decisions) and the final verdict; the document awaits the user's merge review.
Notes
- The writer works in an isolated worktree — the document is only applied after the user reviews and merges it.
- Pass the document path/content forward between rounds so each subagent has the context it needs (summaries, not transcripts).
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.
- 6d ago First seen · 32 lines · 22 tokens per session scan A 8f7dcee6ffd8
Design doc (write -> review loop) is a skill published in the GitHub repository AuraVixStudio/caelo (23 stars, last pushed 24d ago), licensed Apache-2.0. It adds 22 tokens to every session and 369 once invoked, about $0.0001 per session on Opus 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
skill-creator
Use when creating, updating, packaging, or validating an Lfcode skill. Covers user-managed and built-in bundled skills. 用户提到创建 Skill、更新 Skill、打包 Skill、校验 Skill 或技能规范时使用。.
tavern-management
Use when the user asks to import or translate an Lfcode Tavern or SillyTavern character card or worldbook, including PNG image cards, or manage the Tavern plugin lifecycle.
error-collection
记录已验证的工程错误、根因、错误模式和防复发验收规则，作为反面教材使用。.
playwright-browser
Use when controlling browser pages with Playwright MCP in Lfcode; guides navigation, screenshots, interaction, and reuse of hidden or collapsed embedded side browser targets instead of the Electron app shell.
contract-test-engineer
Use when the user asks to verify compatibility between an API provider and consumer, SDK and service, event producer and consumer, schema versions, or serialized data. Define and test the shared contract, including failures and compatibility rules, before changing either side.
e2e-test-engineer
Use when the user asks to create, improve, review, or repair end-to-end tests for a meaningful user workflow across the UI, runtime, backend, or installed application. Validate the real workflow at the highest useful boundary while keeping environments, accounts, and test data controlled.