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 ohong/agent-skills --skill distill-tastegit clone --depth 1 https://github.com/ohong/agent-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/ohong/agent-skills/distill-taste)<a href="https://agentmods.dev/skills/ohong/agent-skills/distill-taste"><img src="https://agentmods.dev/badge/skills/ohong/agent-skills/distill-taste/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/ohong/agent-skills/distill-taste"><img src="https://agentmods.dev/badge/skills/ohong/agent-skills/distill-taste.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 | $0.00095 | $0.00879 |
| Opus 5 | $0.00048 | $0.00439 |
| Sonnet 5 | $0.00019 | $0.00176 |
| Haiku 4.5 | $0.00010 | $0.00088 |
Grade A, and why
distill-taste 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 11d 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Distill Taste
Turn a person's public body of work into an installable ask- skill: a full-roleplay advisor (think Delphi.ai) the user can chat with or hand work to for feedback grounded in that person's actual taste — worldview, quality bar, problem selection, pet peeves, style. Taste has no objectively correct answer; the whole value is fidelity to this person's particular judgment.
Parse the request
<name>— whole-person distillation.<name> on <topic>— still distill the whole person, but weight research and the profile heavily toward that domain.<name> + <name> [+ ...]— a panel: one skill, each person keeping a distinct voice; disagreement between them is signal, never smoothed over.- Rerunning an existing persona refreshes it in place (re-research, rewrite profiles, bump version, re-zip).
If the name is ambiguous (multiple public figures), resolve from context; ask only if genuinely unresolvable.
Research (deep, canonical-first)
Follow references/research-playbook.md. In short: identify the person, then hunt their meta-writing first — advice essays, "how I work/write/decide" pieces, interviews, podcast transcripts, talks, AMAs — before their general output. Read the top ~10–20 primary sources properly (fetch and read, don't skim search snippets). Capture short verbatim quotes with source URLs as you go; these become the persona's citable spine.
Run autonomously — no mid-run checkpoint. Exception — thin data: if you can't find enough substantive primary material for the persona to be more than guesswork, stop, present what you found, and ask whether to proceed with heavy extrapolation, narrow to their one strong domain, or abort.
Build the persona skill
Generate a standalone plugin at the repo root named ask-<kebab-name> (panels: ask-<name1>-<name2>), following references/persona-skill-template.md exactly. Structure mirrors this repo's conventions:
ask-<person>/
.claude-plugin/plugin.json
skills/ask-<person>/
SKILL.md # thin: behavior contract + routing to references
agents/openai.yaml
references/
taste-profile.md # the distilled taste (per person, for panels)
voice.md # how they sound, argue, structure, joke
sources.md # annotated bibliography with quotes + URLs
ask-<person>.skill # zip of skills/ask-<person>/ contents
What ships with it
3 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.
- 11d ago First seen · 49 lines · 95 tokens per session scan A 85fea58f7bd7
distill-taste is a skill published in the GitHub repository ohong/agent-skills (3 stars, last pushed 11d ago), licensed MIT. It adds 95 tokens to every session and 879 once invoked, about $0.0005 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…