telegram-toolkit
Skill191 starsChanged 5 months agoMITView on GitHub
Bot commands, inline keyboards, large file handling, and channel-specific sales strategies for Telegram-based B2B SDR. Register these commands with @BotFather using /setcommands: Step 1 — Need (Product): Step 2 — Budget (Volume): Step 3 — Timeline: Step 4 — Authority: After 3 keyboard interactions, ask naturally in conversation: "Are you the purchasing decision-maker, or should I prepare materials for your team?" (Don't use a keyboard for this — it feels too transactional.)
In this repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- workspace/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- b2b-sdr-agentSkill
- chroma-memorySkill
- delivery-queueSkill
- graphifySkill
- lead-discoverySkill
- quotation-generatorSkill
- sdr-humanizerSkill
- supermemorySkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. About 0 are common, 13 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈13
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 35
Create CRM record: source = telegram_organic, status = new
- Distinctive: line 40
Check CRM for customer's product interest (if returning)
- Distinctive: line 41
If known interest: Send relevant product section + full catalog link
- Distinctive: line 42
If unknown: Send inline keyboard with product categories
- Distinctive: line 43
Always include: specs, MOQ, typical lead time
- Distinctive: line 47
Check if BANT data exists in memory
- Distinctive: line 48
If incomplete: Trigger inline keyboard qualification flow
- Distinctive: line 49
If complete: Generate quote draft → send to owner for approval
- Distinctive: line 53
Read CRM for customer's active records
- Distinctive: line 55
If quote_sent: "Your quote was sent on [date].
- Distinctive: line 56
If no records: "I don't have an active order for you yet.
- Distinctive: line 108
After 3 keyboard interactions, ask naturally in conversation: "Are you the purchasing decision-maker, or should I prepare materials for …
- Distinctive: line 141
When customer is on WhatsApp and needs a large file:
- Very specific: line 156
Detection: Check CRM country field. If Russia/CIS/Iran/Eastern Europe, default to Telegram-first strategy.
- Distinctive: line 177
Never: Direct sales pitches (keep those in DM)
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
telegram-toolkit — Telegram SDR Best Practices & Templates
Bot commands, inline keyboards, large file handling, and channel-specific sales strategies for Telegram-based B2B SDR.
Why Telegram for B2B Sales
| Advantage | Impact |
| No 72h window | Proactive outreach anytime — nurture, follow-ups, stalled leads |
| 2GB file limit | Full product catalogs, certifications, video demos |
| Bot commands | Structured self-service (/catalog, /quote, /status) |
| Inline keyboards | One-tap BANT qualification, 3-5x faster than free-text |
| Username-based | Lower barrier — customer doesn't expose phone number |
| Free API | No per-message cost |
| No account bans | Bot API is stable, unlike WhatsApp's aggressive anti-automation |
Bot Commands
Register these commands with @BotFather using /setcommands:
start - Welcome message and product overview
catalog - Browse product catalog
quote - Request a quotation
status - Check order or quote status
contact - Speak with a human representative
language - Change conversation languageCommand Behavior
/start
- Detect language from Telegram user profile
- Send welcome message with company intro (2-3 sentences max)
- Create CRM record: source =
telegram_organic, status =new - Offer product categories via inline keyboard
- Begin BANT qualification naturally
/catalog
- Check CRM for customer's product interest (if returning)
- If known interest: Send relevant product section + full catalog link
- If unknown: Send inline keyboard with product categories
- Always include: specs, MOQ, typical lead time
- File format: PDF preferred, under 20MB per file
/quote
- Check if BANT data exists in memory
- If incomplete: Trigger inline keyboard qualification flow
- If complete: Generate quote draft → send to owner for approval
- Confirm to customer: "I'm preparing your quotation, will have it ready shortly."
/status
- Read CRM for customer's active records
- Return: latest status, pending actions, next follow-up date
- If quote_sent: "Your quote was sent on [date]. Would you like to discuss it?"
- If no records: "I don't have an active order for you yet. Would you like to start one?"
Inline Keyboard Flows
Quick BANT Qualification
Step 1 — Need (Product):
{
"text": "What products are you interested in?",
"reply_markup": {
"inline_keyboard": [
[{"text": "{{product_1}}", "callback_data": "product_1"}],
[{"text": "{{product_2}}", "callback_data": "product_2"}],
[{"text": "{{product_3}}", "callback_data": "product_3"}],
[{"text": "📋 Full catalog", "callback_data": "full_catalog"}]
]
}
}Step 2 — Budget (Volume):
{
"text": "What's your estimated order quantity?",
"reply_markup": {
"inline_keyboard": [
[{"text": "< 100 units", "callback_data": "qty_small"}],
[{"text": "100-500", "callback_data": "qty_medium"}],
[{"text": "500-1000", "callback_data": "qty_large"}],
[{"text": "1000+", "callback_data": "qty_bulk"}]
]
}
}Step 3 — Timeline:
{
"text": "When do you need delivery?",
"reply_markup": {
"inline_keyboard": [
[{"text": "This month", "callback_data": "timeline_urgent"}],
[{"text": "1-3 months", "callback_data": "timeline_soon"}],
[{"text": "3-6 months", "callback_data": "timeline_planning"}],
[{"text": "Just exploring", "callback_data": "timeline_exploring"}]
]
}
}Step 4 — Authority: After 3 keyboard interactions, ask naturally in conversation: "Are you the purchasing decision-maker, or should I prepare materials for your team?" (Don't use a keyboard for this — it feels too transactional.)
Quick Actions Keyboard
Send after qualification is complete:
{
"text": "How can I help you next?",
"reply_markup": {
"inline_keyboard": [
[{"text": "📋 Get a quote", "callback_data": "action_quote"}],
[{"text": "📦 Product specs", "callback_data": "action_specs"}],
[{"text": "🏭 Factory info", "callback_data": "action_factory"}],
[{"text": "👤 Talk to sales rep", "callback_data": "action_human"}]
]
}
}Large File Strategy
Telegram's 2GB limit makes it the best channel for heavy files:
| Use Case | File | Action |
| Product catalog | PDF, 10-100MB | Send directly via Telegram |
| Certification docs (ISO, CE, etc.) | PDF, 1-20MB | Send on request |
| Product video / factory tour | MP4, 50MB-2GB | Send via Telegram, link on WhatsApp |
| Test reports | PDF, 1-10MB | Send on request |
| Proforma invoice | PDF, < 5MB | Send here + email for formal record |
Cross-channel file routing: When customer is on WhatsApp and needs a large file:
"The full catalog is 85MB — I'll send it to you on Telegram. What's your Telegram username?"
Telegram-First Markets
In these markets, treat Telegram as the primary channel:
| Market | Why Telegram First |
| Russia / CIS | 80%+ business messaging on Telegram |
| Iran | Telegram is the dominant platform |
| Eastern Europe | Strong Telegram adoption for B2B |
| Central Asia | Telegram preferred over WhatsApp |
| Tech/crypto industry | Global preference for Telegram |
Detection: Check CRM country field. If Russia/CIS/Iran/Eastern Europe, default to Telegram-first strategy.
Nurture via Telegram
Telegram has no messaging window — ideal for long-term nurture:
Nurture Cadence (Telegram)
| Timing | Content |
| Day 0 | Initial contact + product overview |
| Day 3 | Relevant case study or industry insight |
| Day 7 | Specific product recommendation based on their interest |
| Day 14 | New product announcement or limited offer |
| Day 30 | Market update or trade show invitation |
| Day 60+ | Quarterly check-in with personalized industry news |
Telegram Channel (One-to-Many)
For customers who follow your brand channel:
- Weekly: Industry news, market trends
- Bi-weekly: New product announcements
- Monthly: Case studies, customer success stories
- Never: Direct sales pitches (keep those in DM)
Security Notes
- Bot Token stored in
secrets.sh, never in config.sh or workspace files dmPolicy: "pairing"requires pairing code — use for exclusive/VIP accessdmPolicy: "open"(not available in all OpenClaw versions) — accepts all DMs- Admin commands restricted to whitelist (same as WhatsApp)
- Rate limit: Same anti-abuse measures as WhatsApp (15 msg/5min, 50 msg/1hr)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

