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 GeckoVision/gecko-surf --skill txline_sharp_agentgit clone --depth 1 https://github.com/GeckoVision/gecko-surfWrote 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/geckovision/gecko-surf/txline_sharp_agent)<a href="https://agentmods.dev/skills/geckovision/gecko-surf/txline_sharp_agent"><img src="https://agentmods.dev/badge/skills/geckovision/gecko-surf/txline_sharp_agent.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 5.1 | $0.00092 | $0.00693 |
| Opus 5 | $0.00046 | $0.00347 |
| Sonnet 5 | $0.00018 | $0.00139 |
| Haiku 4.5 | $0.00009 | $0.00069 |
Grade A, and why
txline-sharp-agent 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 8d 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TxLINE Sharp Movement Detector (powered by Gecko)
Overview
TxLINE is exactly the kind of API agents fail on: paywalled, with a multi-step on-chain auth handshake (guest JWT → on-chain subscribe → sign → activate → two tokens). Gecko comprehends it into first-call-correct tools and injects the credentials invisibly, so your agent just reads odds — and this skill turns that stream into sharp-move signals.
The whole skill runs $0 in recorded mode (synthetic, offline). Live data needs a TxLINE
subscription (see SETUP.md) — a wallet action only the user can take.
Setup
- Comprehend + wire TxLINE (one line, no Python):
npx @geckovision/gecko add \ examples/txline_demo/spec/txline_openapi.yaml \ --base-url https://txline.txodds.com --mode recorded - (Live only) seal the TxLINE session —
SETUP.mdwalks the on-chain subscription. Recorded mode needs nothing.
Workflow
- Detect intent — is the user building a signal tool, a market maker, or a settlement resolver? All three start from the same odds stream.
- Read the feed — call the allow-listed odds tools (
getApiFixturesSnapshot,getApiOddsSnapshotFixtureid,getApiOddsUpdatesFixtureid). Never call the guest/purchase/ activate operations — those are not the agent's job. - Flag sharp moves — run successive snapshots through
SharpDetector(threshold on the implied-probabilityPctdelta). Present each: fixture, book, market, outcome, Δpp, direction. - Reason — for a flagged move, give the single most likely read (steam / injury / lineup / in-play event). Never invent numbers a tool didn't return.
- (Optional) settle on-chain — hand off to
../txodds_settlement: pull TxLINE's Merkle proof and settle a prediction escrow trustlessly. Simulate the tx on a Surfpool mainnet-fork first; a real broadcast is the user's own signed action.
Non-negotiables
- Default to recorded mode ($0) — never flip to live without the user's TxLINE subscription.
- Never expose or call the auth operations; the agent only reads odds.
- Never sign or broadcast a mainnet transaction on the user's behalf. Simulate; hand over.
- Present odds honestly — a flagged move is a signal, not a guarantee.
What ships with it
24 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.
- __init__.py 450 B runs code
- .claude/agents/defi-engineer.md 12 KB
- .claude/agents/NOTICE.md 1.7 KB
- .claude/agents/solana-architect.md 16 KB
- .claude/agents/solana-qa-engineer.md 7.2 KB
- .mcp.json 438 B
- agent.py 3.1 KB runs code
- demo.py 3.3 KB runs code
- detector.py 4.2 KB runs code
- README.md 6.2 KB
- settlement_sim.py 3.8 KB runs code
- SETUP.md 2.1 KB
- story.py 4.6 KB runs code
- surfcall_tools.py 4.1 KB runs code
- synthetic.py 3.4 KB runs code
- tests/__init__.py 0 B runs code
- tests/test_agent.py 2.1 KB runs code
- tests/test_detector.py 2.5 KB runs code
- tests/test_first_call_correct.py 2.0 KB runs code
- tests/test_settlement_sim.py 1.0 KB runs code
- tests/test_story.py 253 B runs code
- tests/test_synthetic.py 1015 B runs code
- tests/test_wallet_sim.py 1.8 KB runs code
- wallet_sim.py 6.4 KB runs code
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.
- 8d ago First seen · 56 lines · 92 tokens per session scan A d49928938ee1
txline-sharp-agent is a skill published in the GitHub repository GeckoVision/gecko-surf (6 stars, last pushed today), licensed Apache-2.0. It adds 92 tokens to every session and 693 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
eastmoney-stock-valuation
A lookup for daily valuation figures on Chinese A-share stocks, such as price-to-earnings, price-to-book, market value and closing price. It can return one stock, a date range, one trading day, or the whole market.
eastmoney-us-stock-daily-ohlc
A lookup for historical daily open, high, low and close prices of a US stock from Eastmoney. With dates, it retrieves the range in small batches and combines the results.
etf-prices
A minute-by-minute price history for one ETF, including price, volume, turnover, average price, and timestamp. An ETF is a fund traded on a stock exchange like a share.
hsi-daily-weight
A market-data lookup for the percentage weights of stocks in Hang Seng family indices. The weight shows how much each constituent stock contributes to an index.
member-build-process
A timeline of one futures member's positions in a chosen contract, combined with estimated daily and cumulative profit or loss based on contract prices.
pledge-detail
A paginated history of share pledges for one listed A-share stock. It includes reporting dates, pledge ratios, pledged share counts, market value, and related share details.