Traadence's Telegram Crypto Signal Bot is a finished signal-delivery system for paid communities that need repeatable screening rather than discretionary posting. It pulls technical, derivatives, on-chain, dominance, wallet-flow, options, and news inputs; rejects unsupported chart patterns; calculates risk-to-reward and expected value; ranks qualified setups into two tracks; and publishes a fixed Telegram message structure on schedule or through exact commands. It does not place orders or independently backtest the published historical rates.
One market scan, two ranked tracks, no invented probabilities, and no directive trade language.
What the Signal Engine Publishes
The primary feed comes from the altFINS analytical API. Each hourly cycle iterates the full Chart Patterns response, accepts only five exact pattern names, and discards every other result. The historical percentage is joined from a Make data store rather than inferred from live data. The published altFINS chart-pattern benchmark supplies the static values used in version one.
| Accepted pattern | Historical rate | Routing rule |
|---|---|---|
| Inverse Head and Shoulders | 84% | Eligible for probability and expected-value ranking |
| Head and Shoulders | 82% | Eligible only on an exact pattern-name match |
| Double Bottom | 82% | Eligible only on an exact pattern-name match |
| Channel Up | 73% | Eligible only on an exact pattern-name match |
| Channel Down | 72% | Eligible only on an exact pattern-name match |
For each accepted setup, the engine calculates risk_to_reward = (target - entry) / (entry - stop) and expected_value = (win_rate × risk_to_reward) - (1 - win_rate). Track A sorts by historical rate; Track B sorts by expected value and always includes “Lower win rate, larger payoff when correct.” A setup selected for both tracks appears only in Track A.
- Header: Track A, Highest Probability, or Track B, Highest EV.
- Setup line: coin, exact pattern name, and timeframe.
- Narrative line: first-person entry watch, personal stop marker, and multiple scale-out targets.
- Statistics line: historical rate, risk-to-reward ratio, and expected value.
- Risk line: “Risking 2 percent of book on this, position size scales to your own account.”
- Context line: optional funding, on-chain, wallet-flow, options, dominance, or news summary.
Core Features
| Feature | Description |
|---|---|
| Exact Pattern Gate | Unmatched or loosely named patterns create misleading statistics. The bot accepts only the five configured altFINS names and drops everything else without guessing or manual-review queues. |
| Dual Ranking Tracks | A single leaderboard hides the difference between probability and payoff. The engine separately ranks historical success rate and calculated expected value, then removes cross-track duplicates. |
| Fixed Risk Disclosure | Account-size assumptions can produce unsafe dollar sizing. Every setup carries the exact 2% book-risk sentence, with no balance field, monetary size, or account-specific calculation. |
| First-Person Message Guard | Directive wording can make an informational post read like an instruction. The Claude Messages API system prompt requires personal-watch language for entry, stop, and multi-target scale-out levels. |
| Scheduled and Command Delivery | Members should not wait for manual posting. The bot publishes at 08:00 and 16:00 UK time and answers only /setup COIN and /watchlist. |
| Freshness and API Fallbacks | Stale or partial market data can corrupt ranking. Each source response is timestamped, validated, and either excluded or marked unavailable before a post is assembled. |
| Outcome Ledger | Without a durable record, later track-record analysis becomes guesswork. Every posted setup stores its source values, message, status, and eventual target, stop, or open outcome. |
Multi-API Architecture
The orchestration runs through Make HTTP modules, scheduled scenarios, routers, iterators, error handlers, and data stores. Telegram's Bot API handles group posts and command replies. Credentials stay in Make connections or secure key storage so a token can be rotated without editing every module.
| Source | Role in the finished workflow |
|---|---|
| CoinGlass API | Open interest, funding, and liquidation context |
| Arkham API | Wallet labels, entity attribution, and whale or exchange movements |
| altFINS | Chart-pattern discovery plus fallback entry, stop, and target levels |
| Deribit API | Options-market context and instrument statistics |
| CoinMarketCap API | Market dominance and rotation context |
| Glassnode API | NUPL, exchange netflow, and on-chain sentiment |
| OpenAI model API | Existing RSS and news-sentiment digest input |
| Claude Messages API | Final structured narrative with enforced disclosure language |
The system treats context feeds as supporting evidence, not as substitutes for the pattern gate. Broader market conditions can change quickly; the CoinGecko annual industry report and Glassnode market-trends research are useful external benchmarks for why derivatives, dominance, and on-chain inputs should be timestamped rather than treated as permanent facts.
Use Cases
- Publish two daily watchlists to a private research group without manually collecting chart, derivatives, wallet-flow, and sentiment data.
- Answer a member’s
/setup SOLrequest with the latest qualified SOL setup or a clear no-active-setup response from the last completed pull. - Keep probability-led ideas separate from larger-payoff ideas while preventing the same coin-pattern setup from appearing twice in one post.
- Maintain a structured outcome history for future reporting without changing the version-one signal-selection rules.
Reliability and Audit Controls
- Source isolation: one failed or rate-limited context API does not silently populate zeros; its field is omitted and the error is logged.
- Staleness rejection: setup timestamps must belong to the latest hourly collection window before ranking or Telegram delivery.
- No padding: each track publishes zero to three qualified setups. The workflow never fills empty slots with rejected patterns.
- Command whitelist: unmatched messages receive “Try
/setupand a coin name, or/watchlist.” rather than open-ended trading interpretation. - Traceable calculations: stored inputs preserve entry, stop, targets, rate, ratio, expected value, ranking position, and post identifier.
Traadence also handles signal-system integration and ongoing maintenance when posting schedules, API adapters, pattern tables, or monitoring rules need controlled changes after deployment.
Project Directory
alpha-inner-circle-bot/
├── scenarios/
│ ├── 01_hourly_data_pull.blueprint.json
│ ├── 02_rank_and_format.blueprint.json
│ ├── 03_scheduled_posts.blueprint.json
│ ├── 04_command_responder.blueprint.json
│ └── 05_outcome_tracker.blueprint.json
├── config/
│ ├── pattern_win_rates.json
│ ├── message_templates.json
│ ├── command_rules.json
│ └── staleness_policy.json
├── prompts/
│ ├── claude_system_prompt.md
│ ├── scheduled_post_prompt.md
│ └── setup_reply_prompt.md
├── docs/
│ ├── scenario-map.md
│ ├── credential-rotation.md
│ ├── operations-runbook.md
│ └── test-evidence.md
└── README.md
Operational Benchmarks
| Acceptance check | Required behavior |
|---|---|
| Collection cadence | Run every 60 minutes on the hour |
| Scheduled delivery | Post at 08:00 and 16:00 UK time daily |
| Track capacity | Maximum three setups per track; zero is valid |
| Duplicate control | One setup may appear in only one track per cycle |
| Command coverage | Live tests for /setup COIN, /watchlist, and fallback text |
| Delivery proof | At least one complete scheduled cycle recorded end to end |
How to Publish Ranked Setups Using Traadence's Telegram Crypto Signal Bot
Download & Set Up the Project
Download, set up, and install Traadence's Telegram Crypto Signal Bot to get the project running. If you hit any difficulty, contact us here.
Open the Make Scenarios
Open the organized scenario folder, verify active connections, and confirm the hourly collector, ranking flow, scheduler, command responder, and outcome tracker are enabled.
Confirm Rules and Delivery
Review the five-pattern data store, UK posting times, Telegram chat ID, 2% risk wording, freshness window, and Claude message template before running.
Run and Verify Outputs
Trigger Run once, inspect both ranked tracks, then test /setup SOL and /watchlist; results appear in Telegram and the outcome ledger.
Questions
How do you create a Telegram bot for crypto trading signals?
Create the bot token in BotFather, connect it to Make through Telegram’s API, and separate the workflow into collection, filtering, ranking, formatting, delivery, and logging stages. This build also uses an exact-pattern data store, deterministic calculations, command whitelisting, freshness checks, and a fixed language prompt so scheduled posts and replies follow the same rules.