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 SunflowersLwtech/sunkit --skill search-harnessgit clone --depth 1 https://github.com/SunflowersLwtech/sunkitWrote 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/sunflowerslwtech/sunkit/search-harness)<a href="https://agentmods.dev/skills/sunflowerslwtech/sunkit/search-harness"><img src="https://agentmods.dev/badge/skills/sunflowerslwtech/sunkit/search-harness/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/sunflowerslwtech/sunkit/search-harness"><img src="https://agentmods.dev/badge/skills/sunflowerslwtech/sunkit/search-harness.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 | $0.00060 | $0.00305 |
| Opus 5.5 | $0.00024 | $0.00122 |
| Sonnet 5.5 | $0.00012 | $0.00061 |
| Haiku 4.5 | $0.00006 | $0.00030 |
Grade A, and why
search-harness 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 2mo 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.
What it actually says
Search Harness — Domain Knowledge Pack
This skill packages the scoping and research documents for a search harness for coding agents (CLI + skills design). No code yet — this is the reference layer for decisions, reviews, and onboarding.
How to use these documents
Read the file that matches the question, in this order of relevance:
| Question | File |
|---|---|
| What is this domain? Terms, players, accumulated conclusions | domain-knowledge.md |
| What are we building and in what order? | roadmap.md |
| Which sources/APIs are in scope? | sources.md |
| How do we wrap a ground-truth search API? | ground-api-pipeline.md |
| Why that strategy? | ground-api-strategy.md |
| Raw API documentation for candidate sources | api-docs/ |
Ground rules carried over from the project
- Domain knowledge is a reference manual: consult it before re-deriving conclusions that are already recorded.
- The roadmap's top priority is "equip every source" — check
roadmap.mdbefore proposing new scope.
What ships with it
34 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.
- .gitignore 934 B
- api-docs/01-github.md 55 KB
- api-docs/02-gitlab.md 18 KB
- api-docs/03-stackexchange.md 2.8 KB
- api-docs/04-hn-algolia.md 4.6 KB
- api-docs/05-reddit.md 72 KB
- api-docs/06-arxiv.md 40 KB
- api-docs/07-openalex.md 13 KB
- api-docs/08-openreview.md 6.2 KB
- api-docs/08c-openreview-py.md 8.1 KB
- api-docs/09-libraries-io.md 3112 KB
- api-docs/10-huggingface.md 2.0 KB
- api-docs/11-nvd.md 49 KB
- api-docs/12-twitterapi.md 2.2 KB
- api-docs/13-osv.md 1.2 KB
- api-docs/13b-osv-query.md 6.7 KB
- api-docs/14-deps-dev.md 56 KB
- api-docs/15-lobsters.md 9.2 KB
- api-docs/16-bluesky.md 74 KB
- api-docs/17-sourcegraph.md 8.6 KB
- api-docs/18-exa.md 9.9 KB
- api-docs/19-context7.md 6.3 KB
- api-docs/20-playwright.md 22 KB
- api-docs/21-deepwiki.md 2.2 KB
- api-docs/22-yt-dlp.md 170 KB
- api-docs/23-jina-reader.md 28 KB
- api-docs/24-firecrawl.md 12 KB
- api-docs/INDEX.md 4.0 KB
- domain-knowledge.md 145 KB
- ground-api-pipeline.md 13 KB
- ground-api-strategy.md 11 KB
- index.html 68 KB
- roadmap.md 12 KB
- sources.md 7.5 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.
- 2mo ago First seen · 31 lines · 60 tokens per session scan A 41cb14be6361
search-harness is a skill published in the GitHub repository SunflowersLwtech/sunkit (13 stars, last pushed yesterday), licensed Apache-2.0. It adds 60 tokens to every session and 305 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
api-docs-generator
Generate structured API reference documentation from source code, OpenAPI specs, or route definitions. Trigger phrases include "generate API docs", "document this API", "create API reference", "write endpoint docs", "scaffold API docs".
api-tester
A tool for creating and checking API tests from the real API contract and implementation. An API is the agreed way that software sends requests and receives responses.
nextjs-supabase-vercel
Set up the current empty folder with the AI App Starter, so the agent builds a Next.js, Supabase and Vercel app with the owner step by step, from idea to launch.
doku-payment-gateway
Expert guide for integrating DOKU Payment Gateway: DOKU Checkout (Hosted & Modal Popup with HMAC-SHA256), Direct API SNAP BI Standard (VA, QRIS, Direct Debit with HMAC-SHA512), and Direct API Non-SNAP (Card Payment Page, OVO Push). Covers single-integration checkout, direct host-to-host APIs, webhook verification, and…
email-notification-expert
Expert guide for transactional email (Resend, Postmark, SES), React Email templates, in-app notifications, and unified communication pipelines / Panduan ahli untuk email transaksional (Resend, Postmark, SES), template React Email, notifikasi in-app, dan pipeline komunikasi terpadu.
file-upload-media-expert
Expert guide for file uploads (S3, R2, Supabase Storage), presigned URLs, image/video processing, CDN optimization, and media pipeline architecture / Panduan ahli untuk upload file (S3, R2, Supabase Storage), presigned URL, pemrosesan gambar/video, optimasi CDN, dan arsitektur pipeline media.