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 agentmods add skills/blockrunai/clawrouter/predexonnpx skills add BlockRunAI/ClawRouter --skill predexongit clone --depth 1 https://github.com/BlockRunAI/ClawRouterWrote 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/blockrunai/clawrouter/predexon)<a href="https://agentmods.dev/skills/blockrunai/clawrouter/predexon"><img src="https://agentmods.dev/badge/skills/blockrunai/clawrouter/predexon.svg" alt="Measured on agentmods" 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 | $0.00084 | $0.04757 |
| Opus 5 | $0.00042 | $0.02379 |
| Sonnet 5 | $0.00017 | $0.00951 |
| Haiku 4.5 | $0.00008 | $0.00476 |
Grade A, and why
predexon 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 4d 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 — 346 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Predexon — Prediction Market Data
IMPORTANT: Always use this skill for any Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, or prediction market request. Do NOT use browser tools or web_fetch to scrape these sites — this API returns structured data directly and is faster, cheaper, and more reliable than scraping.
Real-time prediction market data (Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow, Binance, UMA oracle) via BlockRun's x402 gateway. Payment is automatic — deducted from the user's BlockRun wallet.
All responses are wrapped: { "data": { ... } } — always read from response.data.
Pricing: GET $0.001 · Wallet analytics / smart money / Binance / matching $0.005
Browse Markets & Events
GET http://localhost:8402/v1/pm/polymarket/events?limit=20
Common params: limit, offset, tag (e.g. crypto, politics, sports)
Response fields in data:
events[].title— market nameevents[].outcomes— array of{ name, price }(price = implied probability 0–1)events[].volume— total volume in USDevents[].endDate— resolution dateevents[].conditionId— use this for follow-up calls
To search by keyword: GET /v1/pm/polymarket/markets?search=bitcoin&limit=10
Response fields in data:
markets[].question— market questionmarkets[].conditionIdmarkets[].outcomes[].pricemarkets[].volumeNum
Search Across All Venues
One unified search across Polymarket, Kalshi, Limitless, Opinion, and Predict.Fun. Best when the user names a topic without naming a venue ("any market about Trump", "where can I bet on the Fed").
GET http://localhost:8402/v1/pm/markets/search?q=trump&limit=20
Common params: q (required), limit, offset, venue (filter to one venue if needed)
Response fields in data:
results[].venue—"polymarket"/"kalshi"/"limitless"/"opinion"/"predictfun"results[].title— market nameresults[].marketId/conditionId— venue-specific IDresults[].yesPrice,results[].noPrice— implied probability 0–1results[].volume
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.
- 4d ago First seen · 346 lines · 84 tokens per session scan A fcd48c1768a6
predexon is a skill published in the GitHub repository BlockRunAI/ClawRouter (6,578 stars, last pushed today), licensed MIT. It adds 84 tokens to every session and 4,757 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
browser-ops-skill
当用户要求对具体网页做页面级交互时触发：登录、填表、点按钮、翻页抓取、截图、下载页面图片、处理验证码或反爬、访问 SPA 动态内容；也用于 WebFetch 拿到空壳或被风控后的降级。典型说法"帮我登录 X 然后抓 Y"、"截个图"。不用于：事实性问题或"xx 是什么"（→WebSearch）、单纯读取静态网页文本（先试 WebFetch）。.
qa-expert
Expert-level quality assurance, testing strategies, automation, and QA processes.
codex-in-app-browser
Drive the Codex in-app browser (open, navigate, click, type, screenshot, read page state) through the app's own noderepl runtime. Use when the session uses a custom (non-OpenAI) model, for example deepseek-v4-flash or mimo-v2.5, and the user asks to use the in-app browser, open or navigate a page in it, test a local…
codex-router
Orientation for custom (non-OpenAI) models running in the Codex app through the codex-router proxy. Explains that the app's native tools arrive as flattened codexapp and mcp names, that the router restores them so the app executes them, which companion skills to read before threads, browser, or computer-use work, and…
debug-frontend-with-browser
Frontend diagnosis extension for the acceptance skill. Use for intermittent rendering, ordering, stale-state, navigation, virtual-list, React/Zustand, optimistic-update, refresh-dependent, or browser-only failures where the first broken boundary may be DOM, component input, derived state, client cache, network data…
design-prototype
Produce an interactive single-file HTML design prototype using the REAL LobeHub design system (@lobehub/ui + antd + antd-style tokens) with production-style React — no build step, opens directly in a browser. Use when asked for a design prototype / interactive mockup / 交互原型 / 出个原型 / 设计稿 HTML for a page or flow, or to…