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 MariHQ/mari-cc --skill marigit clone --depth 1 https://github.com/MariHQ/mari-ccWrote 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/marihq/mari-cc/mari)<a href="https://agentmods.dev/skills/marihq/mari-cc/mari"><img src="https://agentmods.dev/badge/skills/marihq/mari-cc/mari/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/marihq/mari-cc/mari"><img src="https://agentmods.dev/badge/skills/marihq/mari-cc/mari.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 | $0.00072 | $0.00738 |
| Opus 5 | $0.00036 | $0.00369 |
| Sonnet 5 | $0.00014 | $0.00148 |
| Haiku 4.5 | $0.00007 | $0.00074 |
Grade A, and why
mari 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 9d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mari
Mari is a prose and documentation system. Its detector supplies concrete findings; you make the editorial judgment and edits.
Before editing
- Read the target and at least one representative file from the same project.
- Read
PRODUCT.mdandSTYLE.mdwhen present. - Choose the closest register reference: docs, marketing, editorial, or microcopy.
- Run
mari detect <target>and use the highest-value findings to guide the pass. - For a recognized document type, run
mari asset detect <target>, load its reference, and runmari asset check <target>.
Project settings are read from .mari/config.json, with personal repository overrides in .mari/config.local.json.
Routing
deslop,tighten,clarify,sharpen,understate,critique,polish, anddraft: load the matching reference and follow it.voice,cadence,format,delight,harden,adapt,localize, andnarrative: load the matching reference.detect,audit,asset,platform,surface,check,i18n,rules,nudge,glossary,config,status, andfeatures: run the CLI command directly. Use--strictwhen the user asks for a release or CI gate.- A general request such as “make this better”: inspect the text first, then choose the smallest useful pass. Use
deslopfor canned phrasing,tightenfor excess length,clarifyfor ambiguity, andpolishfor a final multi-category pass.
Editing standard
- Preserve facts, names, numbers, links, examples, and constraints unless the user asks to change them.
- Match the project's existing register instead of imposing a generic voice.
- Prefer direct sentences, specific verbs, varied cadence, and concrete claims.
- Remove throat-clearing, repeated conclusions, canned transitions, inflated language, fake quotations, and unnecessary headings.
- Do not optimize for a detector score at the expense of meaning.
- After editing, run
mari detect <target>again and review the diff. - Explain material changes briefly; do not narrate every mechanical edit.
What ships with it
46 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.
- references/reference-adapt.md 987 B
- references/reference-asset-adr.md 1.4 KB
- references/reference-asset-code-of-conduct.md 1.3 KB
- references/reference-asset-contributing.md 1.3 KB
- references/reference-asset-governance.md 1.3 KB
- references/reference-asset-postmortem.md 1.8 KB
- references/reference-asset-rfc.md 1.5 KB
- references/reference-asset-runbook.md 1.4 KB
- references/reference-asset-security.md 1.3 KB
- references/reference-audit.md 835 B
- references/reference-cadence.md 931 B
- references/reference-clarify.md 833 B
- references/reference-community.md 6.5 KB
- references/reference-critique.md 956 B
- references/reference-delight.md 887 B
- references/reference-deslop.md 2.6 KB
- references/reference-document.md 1.5 KB
- references/reference-draft.md 1.0 KB
- references/reference-format.md 1.8 KB
- references/reference-glossary.md 862 B
- references/reference-harden.md 940 B
- references/reference-init.md 447 B
- references/reference-live.md 1.0 KB
- references/reference-localize.md 858 B
- references/reference-narrative.md 6.3 KB
- references/reference-outline.md 962 B
- references/reference-platform.md 4.7 KB
- references/reference-polish.md 948 B
- references/reference-register-docs.md 891 B
- references/reference-register-editorial.md 859 B
- references/reference-register-marketing.md 782 B
- references/reference-register-microcopy.md 857 B
- references/reference-sharpen.md 918 B
- references/reference-soften.md 919 B
- references/reference-tighten.md 852 B
- references/reference-understate.md 2.7 KB
- references/reference-voice.md 991 B
- references/template-adr.md 2.1 KB
- references/template-code-of-conduct.md 2.6 KB
- references/template-contributing.md 3.0 KB
- references/template-governance.md 2.6 KB
- references/template-postmortem.md 2.6 KB
- references/template-readme.md 4.0 KB
- references/template-rfc.md 2.7 KB
- references/template-runbook.md 2.1 KB
- references/template-security.md 2.7 KB
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.
- 9d ago First seen · 51 lines · 72 tokens per session scan A 446761bd6063
mari is a skill published in the GitHub repository MariHQ/mari-cc (32 stars, last pushed 29d ago), licensed MIT. It adds 72 tokens to every session and 738 once invoked, about $0.0004 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
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
azure-ai-translation-text-py
Azure AI Text Translation SDK for real-time text translation, transliteration, language detection, and dictionary lookup. Use for translating text content in applications. Triggers: "text translation", "translator", "translate text", "transliterate", "TextTranslationClient".
harden
Improve interface resilience through better error handling, i18n support, text overflow handling, and edge case management. Makes interfaces robust and production-ready. Use when the user asks to harden, make production-ready, handle edge cases, add error states, or fix overflow and i18n issues.
seedance-vocab-ja
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Japanese.
asc-subscription-localization
Bulk-localize subscription, subscription-group, and in-app purchase display names across App Store locales using asc, including API 4.4.1 version-scoped v2 resources. Use when filling or updating subscription/IAP names and descriptions without App Store Connect UI work.
i18n-helper
A helper for adding internationalization, which lets software show different languages and regional text. It finds user-visible text written directly in code and moves it into language files.