what is voight-txline-agent fr?

voightxyz/voight-txline-agent — explained in plain English

Analysis updated 2026-05-18

★ 11PythonAudience · developerComplexity · 4/5LicenseSetup · moderate

vibe map

mindmap
  root((repo))
    What it does
      Detects sharp odds movements
      Explains cause via research
      Sends Telegram alerts
    Tech stack
      Python
      Solana
      Voight agent platform
    Use cases
      Monitor live sports odds
      Verify detector on recorded data
      Anchor predictions on chain
    Audience
      Sports data researchers
      Agent platform builders
    Design
      Deterministic math script
      LLM explains not calculates
      Auditable ledger

mindmap root((repo)) What it does Detects sharp odds movements Explains cause via research Sends Telegram alerts Tech stack Python Solana Voight agent platform Use cases Monitor live sports odds Verify detector on recorded data Anchor predictions on chain Audience Sports data researchers Agent platform builders Design Deterministic math script LLM explains not calculates Auditable ledger

what do people make with this?

VIBE 1

Run the included recorded data capture to verify the sharp-movement detector locally with no API keys

VIBE 2

Watch a live World Cup fixture and receive a Telegram alert when odds move sharply

VIBE 3

Anchor a prediction ledger entry on Solana so it can be proven to predate the event

VIBE 4

Study the deterministic detection algorithm as a template for other market monitoring agents

what's the stack?

PythonSolanaTelegram Bot APIMetaplex

how it stacks up fr

	voightxyz/voight-txline-agent	2arons/llm-cli	abe238/claude-video-plus
Stars	11	11	11
Language	Python	Python	Python
Setup difficulty	moderate	easy	easy
Complexity	4/5	2/5	3/5
Audience	developer	developer	developer

Figures from each repo's GitHub metadata at analysis time.

how do i run it?

Difficulty · moderate time til it works · 30min

Local replay against the included data capture needs only Python 3.10+, live use requires a free TxLINE API token and, for on-chain attestation, a funded Solana wallet.

MIT license: use, copy, modify, and distribute freely, provided as is with no warranty.

in plain english

voight-txline-agent is a hackathon project built for the TxLINE World Cup Hackathon 2026. It is an autonomous agent that watches live sports odds data from TxLINE, a service that publishes World Cup betting odds anchored on the Solana blockchain, and looks for sharp movements, meaning sudden, unusual shifts in the implied probability of an outcome. When it spots one, it investigates the likely cause using live web research, such as a goal or a red card, sends an alert to its owner over Telegram with a summary card, and logs the event to a ledger that can be audited later. The project deliberately separates two jobs. A plain Python script does the actual math: it builds a probability series from TxLINE's odds feed and flags a movement only when it crosses two thresholds at once, a minimum point change within a short time window and a statistical unusualness check against that market's own recent noise, followed by a cooldown period so the same market does not alert repeatedly. Only after something is flagged does the AI agent step in, to explain what happened and deliver the alert. The README states plainly that the language model never does the market math itself. Anyone can verify the detector's behavior locally without any API keys, using a real recorded data capture from a World Cup semifinal that ships inside the repository, and the README reports the exact result that run should produce. Running against live data requires a free TxLINE API token obtained through a one time on chain transaction. The project also includes a separate tool that anchors each prediction the agent makes onto the Solana blockchain as a hashed, timestamped memo, so a pick can be proven to have existed before the event it predicted, without needing to trust the project's own claims. The deployed version runs as a scheduled cloud agent built on the Voight platform, and the repository is written in Python, released under the MIT license, and requires a Solana wallet only for the optional on chain attestation step.

prompts (copy fr)

prompt 1

Walk me through running the sharp_detect.py replay command against the included semifinal capture.

prompt 2

Explain the two conditions the detector uses to decide a market movement counts as sharp.

prompt 3

Show me how to get a free TxLINE API token so I can run the detector against live odds data.

prompt 4

Help me understand how attest.mjs anchors a prediction on the Solana blockchain.

Frequently asked questions