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/wonsukchoi/domain-expertsWrote 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/commands/wonsukchoi/domain-experts/domain-expert)<a href="https://agentmods.dev/commands/wonsukchoi/domain-experts/domain-expert"><img src="https://agentmods.dev/badge/commands/wonsukchoi/domain-experts/domain-expert/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/commands/wonsukchoi/domain-experts/domain-expert"><img src="https://agentmods.dev/badge/commands/wonsukchoi/domain-experts/domain-expert.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 | $0.00018 | $0.00295 |
| Opus 5 | $0.00009 | $0.00148 |
| Sonnet 5 | $0.00004 | $0.00059 |
| Haiku 4.5 | $0.00002 | $0.00030 |
Grade A, and why
domain-expert 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.
This is a copy
88% identical to domain-expert.codex — 1 line differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Task: $ARGUMENTS
- Run:
npx --yes domain-experts match "$ARGUMENTS" --json - If
confident: true:- Read the matched role's
SKILL.mdin full (path inbest.file), and anyreferences/files the task needs the depth of. - Mention which role you loaded, in one line, before answering.
- Adopt its reasoning for the rest of this response: follow its Decision framework, apply its Mental models & heuristics, match its Communication style.
- If
metadata.maturityisdraft, use it fully but don't claim practitioner review if asked.
- Read the matched role's
- If
confident: false:- State plainly that no role covers this yet. Name the closest low-confidence candidates if any were returned.
- Point at
ROADMAP.mdin the domain-experts repo for the nearest O*NET occupation, and mention a PR perCONTRIBUTING.mdwould add it. - Then answer the task as a generalist if that's clearly still wanted — flagged explicitly as improvised, not sourced from a vetted role.
- Never blend step 2 and step 3 framing — a loaded role's answer and an improvised one must read as visibly different confidence levels.
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 · 18 lines · 18 tokens per session scan A 5e459eccc946
domain-expert is a command published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed yesterday), licensed MIT. It adds 18 tokens to every session and 295 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to domain-expert.codex, differing in 1 line, and is treated as a copy.
Other commands, from other repositories
prisma-optimize
Optimize Prisma queries with automated analysis, N+1 detection, index recommendations, and performance patterns based on official documentation.
design-mobile
Design a mobile screen or flow. This command never writes SwiftUI or Compose — the deliverable is a device-framed HTML mockup plus a handoff spec for the platform developer.
run-tasks
Execute GitHub issues using full EPCT workflow (Explore-Plan-Code-Test). Converts issue requirements into completed PR.
design-page
Routes a new page or screen in an existing project by register and move; taste-first may defer existing tokens until concept lock.
test
Test commit. Use for add tests, update tests, test coverage, unit test, spec.
checklist
Generate a custom checklist for the current feature based on user requirements.