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 po4yka/llm-wiki-skills --skill llm-wiki-human-first-designgit clone --depth 1 https://github.com/po4yka/llm-wiki-skillsWrote 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/po4yka/llm-wiki-skills/llm-wiki-human-first-design)<a href="https://agentmods.dev/skills/po4yka/llm-wiki-skills/llm-wiki-human-first-design"><img src="https://agentmods.dev/badge/skills/po4yka/llm-wiki-skills/llm-wiki-human-first-design/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/po4yka/llm-wiki-skills/llm-wiki-human-first-design"><img src="https://agentmods.dev/badge/skills/po4yka/llm-wiki-skills/llm-wiki-human-first-design.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 | $0.00059 | $0.01424 |
| Opus 5 | $0.00030 | $0.00712 |
| Sonnet 5 | $0.00012 | $0.00285 |
| Haiku 4.5 | $0.00006 | $0.00142 |
Grade A, and why
llm-wiki-human-first-design 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 12d 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 — 236 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM-Wiki Human-First Design
Goal
Help users design or audit an LLM-Wiki as a human-readable shared context layer, not an opaque agent cache or a folder of generated Markdown summaries.
When to use
Use when the user says or asks:
- "LLM-Wiki should not be considered in a vacuum."
- "If agents disappear tomorrow, can humans still find information?"
- "I do not want another Confluence-like document dump."
- "How do we make the wiki readable and useful for humans?"
- "How do we prevent agent-generated markdown from becoming slop?"
- "What page structure makes the wiki navigable?"
- "How do we test human findability?"
- "How do we design a wiki around decisions/questions rather than sources/spaces?"
Inputs
- The current wiki tree (or a proposed layout) — directory structure,
index.md, domain maps. - A handful of recurring questions humans actually ask (or need to derive them if the user has not listed any).
- Sample pages, especially any that mix generated summary with human-authored synthesis.
- Any existing page-contract or frontmatter convention already in use, so the audit can search for deviations instead of inventing a new contract.
- Whether the wiki currently has agent write access, and under what review process — needed to assess the draft/review/promote gate in step 7.
Required references
Read these when available:
references/docs/24-human-first-llm-wiki.mdreferences/docs/adoption-objections.mdreferences/docs/adoption-q-and-a.mdreferences/docs/criticism-and-mitigations.mdreferences/docs/04-anti-slop-and-trust.md
Core stance
Use this answer:
The objection is correct. LLM-Wiki should be human-first and agent-assisted. Humans must be able to find, read and challenge the knowledge without an agent. Agents should maintain bookkeeping, links, indexes, drafts, lint reports and refresh queues; they should not become the owner of truth.
Design principles
human-readable by default
agent-optimized by metadata
source-backed by construction
decision-oriented by structure
What ships with it
5 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.
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.
- 12d ago First seen · 236 lines · 59 tokens per session scan A c5b525c1f922
llm-wiki-human-first-design is a skill published in the GitHub repository po4yka/llm-wiki-skills (3 stars, last pushed 19d ago), licensed MIT. It adds 59 tokens to every session and 1,424 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
mstar-design-md
DESIGN.md design system specification for Morning Star projects. Create, audit, and maintain project-level design tokens (Colors, Typography, Spacing, Elevation, Motion, Shapes, Components, Voice & Content) using Vercel Geist as reference template. Three-level completeness checklist (MVP/Standard/Production) with…
ui-ux-pro-max
UI/UX design intelligence with 50+ styles, 161 palettes, 57 font pairings, 99 UX guidelines, and 25 chart types. Searchable database with priority-based recommendations for web and mobile.
agent-hephaestus
UI, accessibility, and interaction-quality review.
okf-harness-bootstrap
This unified entrypoint routes setup and workspace maintenance without making the user choose a mode.
frontier-status-summary
Maps which assets to use for externally distributed materials (briefings, cards, guides) by audience type. Performs cross-referencing only — does not transform hub asset content.
okf-harness
One Door routes OKF Harness workspace requests to exactly one internal workflow at a time.