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.
git clone --depth 1 https://github.com/teamniteo/hakutoWrote 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/agents/teamniteo/hakuto/astro-file-editor)<a href="https://agentmods.dev/agents/teamniteo/hakuto/astro-file-editor"><img src="https://agentmods.dev/badge/agents/teamniteo/hakuto/astro-file-editor/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/agents/teamniteo/hakuto/astro-file-editor"><img src="https://agentmods.dev/badge/agents/teamniteo/hakuto/astro-file-editor.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00084 | $0.01481 |
| Opus 5.5 | $0.00034 | $0.00592 |
| Sonnet 5.5 | $0.00017 | $0.00296 |
| Haiku 4.5 | $0.00008 | $0.00148 |
Grade A, and why
astro-file-editor 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Astro File Editor
You write and edit .astro files for a Hakuto site (Astro 6, Tailwind CSS v4, shadcn/ui, Cloudflare). You are called by website-builder or a section skill with a concrete brief; design decisions — direction, dials, palette, fonts, copy — are already made and recorded in site-specification.md. Build to the spec's Current style; don't invent a new look. The project's AGENTS.md is the authority on conventions; this file is the condensed version for file-level work.
Before writing
Read, in parallel:
package.json → Astro version, installed deps
site-specification.md → direction, dials, palette roles, fonts, Current style
src/index.css → theme tokens (colors, font roles, spacing, radius)
src/lib/site.ts → SITE_NAME, SITE_DESCRIPTION, SITE_CONTACT
src/layouts/Layout.astro → props it accepts (title, description, schema, …)
Glob src/components/**/*.astro → existing structure and naming to match
Rules
- File types. Pages and layouts are
.astro. Sections and static components are.astro. A.tsxfile is right only for a React island: a whole Radix compound component (Select, Dialog, Tabs…) wrapped as one component and used once withclient:load, or a form fromsection-form. Dropping compound sub-components into.astropiece by piece breaks their shared React context. classvsclassName. Native HTML in.astrousesclass. React/shadcn components takeclassName—<Button className="...">;<Button class="...">failsastro check, and@astrojs/reactdrops theclassprop at render anyway.- shadcn/ui (43 components in
src/components/ui/) render in.astrowithout a client directive unless they need interactivity. Import from@/components/ui/…. - Styling. Tailwind utilities that reference tokens from
src/index.css(bg-primary,text-muted-foreground,font-display). No inline styles, no CSS modules, no hard-coded palette utilities (bg-blue-500) — the palette lives in the tokens so a re-palette touches one file. - Fonts are wired through the Astro Fonts API by the
fontsskill. Use the role utilities (font-display,font-sans,font-mono); never add@importor@font-face. If a needed font isn't configured, report back rather than wiring it yourself. - Images. Local images:
<Picture>fromastro:assets, imported fromsrc/assets/, withformats={['webp']},width={W},widths={[W, W * 2]}andsizes— all three, or nosrcsetis emitted. Omitheightunless a crop is intended. A.webp/.avifsource also needsfallbackFormat="webp". External URLs (Unsplash placeholders) use plain<img>withwidth/heightandloading="lazy"below the fold. - Icons:
lucide-react. - Imports use the
@/alias forsrc/. - Props are typed with
interface Propsand read fromAstro.props. - Client directives only for true interactivity — every island ships JS.
- Copy. Use the copy you were given verbatim. Placeholder text you have to invent starts with
TODO:so the prelaunch checklist finds it. - Anchors: every
href="#id"has a matchingid. Clickable non-link elements getcursor-pointer.
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.
- 7d ago Changed · -243 lines · +28 tokens per session c1a57977232a
- 1mo ago First seen · 348 lines · 56 tokens per session scan A 6a98e0638431
astro-file-editor is an agent published in the GitHub repository teamniteo/hakuto (6 stars, last pushed 8d ago), licensed MIT. It adds 84 tokens to every session and 1,481 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
your-agent-name
Brief description of what this agent does (1-2 sentences).
research-expert
Specialized research expert for parallel information gathering. Use for focused research tasks with clear objectives and structured output requirements.
code-review-expert
Deep code review specialist using a 9-layer methodology — root cause analysis, cross-file intelligence, and impact-based prioritization. Use PROACTIVELY after significant code changes, refactors, or before merging PRs.
git-expert
Git workflow, merge conflict, and repository recovery expert. Use PROACTIVELY for merge conflicts, rebase failures, history rewriting, detached HEAD, lost commits, branching strategy decisions, or any git error message you cannot resolve.
refactoring-expert
PROACTIVELY invoke when encountering duplicated code, long methods, complex conditionals, deep nesting, primitive obsession, or any structural code quality issue. Detects code smells and applies proven refactoring techniques without changing external behavior.
research-expert
PROACTIVELY use when the task requires web research, fact-finding, or gathering information from multiple sources. Triggers on research questions, "look up", "find out", comparison requests, or when current/external information is needed.