O.C. Manjos Lead Qualifier
SkillDatabases & dataLets your agent score incoming WhatsApp sales leads and automatically call promising ones back.
Use O.C. Manjos Lead Qualifier in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add O.C. Manjos Lead Qualifier and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the O.C. Manjos Lead Qualifier skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
About this skill
Qualifies inbound sales enquiries from a Supabase-backed WhatsApp sales AI and places outbound follow-up calls via CALL-E for high-value, recently-active leads.
What this skill tells your AI
The instructions your AI receives, as published by calle-ai/awesome-phone-call-agents in skills/ocmanjos-lead-qualifier/SKILL.md and read by Ahel’s review.
Scores inbound product enquiries (captured via a WhatsApp sales AI into Supabase) against category, price, and recency rules, then places a CALL-E outbound call to qualifying leads to confirm interest and capture delivery or pickup preference.
Use it for
- Following up on high-value product enquiries that went unanswered on WhatsApp within 72 hours
- Confirming customer interest and delivery preference via voice instead of relying on repeat text follow-ups
- Demonstrating a real small-business use case for CALL-E: filtering noisy, free-text enquiry data down to genuinely qualifying leads before spending a call
Do not use it to
- Call numbers not sourced from an explicit customer enquiry
- Call leads marked as already purchased, a known competitor, or opted out
- Guess phone numbers, prices, or product categories - the skill only prices and calls what matches configured rules and real price-sheet data
Setup
- A Supabase table with enquiry records (product_asked, created_at, whatsapp_responded, known_competitor, already_purchased, opted_out)
- A product price sheet (.xlsx) with description, unit price, and sell price columns
- .env with SUPABASE_URL and SUPABASE_ANON_KEY (never commit this file)
- CALL-E CLI authenticated (calle auth login)
How it works
- Pull recent enquiries from Supabase
- Extract a product keyword and variant (e.g. size/phase) from free-text product_asked
- Look up the real sell price from the price sheet, matched by token, not exact phrase
- Score against category/price/recency/disqualifier rules
-
- For qualifying leads, the operator manually runs CALL-E's plan_call via the CLI, reviews the confirm_summary, then runs run_call. This script identifies who to call; it does not itself invoke CALL-E.
Qualification rule (example, tune to your own product line)
IF (category = "Distribution Board" AND value >= 35000) OR (category = "Box/Conduit" AND value >= 2500) OR (category = "Solar/Flood Light" AND value >= 34000) AND enquiry_age <= 72h AND no_whatsapp_response = true AND NOT (known_competitor OR already_purchased OR opted_out) THEN CALL
Side effects
This skill places real outbound phone calls when a lead qualifies. Each call consumes CALL-E call credits. Always review plan_call output (ready_to_run, confirm_summary) before calling run_call - this repository's rules prohibit placing a setup-time test call without the user explicitly asking for one.
Example (masked)
Phone: +15550101234 (masked example - never a real customer number in samples) Product: 3-phase distribution board (D6) Price: 42,500 NGN Result: customer confirmed interest, requested delivery, did not provide delivery details before call ended
Known limitations
- Keyword/variant extraction currently only handles size and phase patterns for the Distribution Board category; other multi-variant product lines need similar extraction logic added
- Phone numbers must currently be sourced manually until a WhatsApp Business API integration writes them into Supabase directly
- Call-outcome confidence scoring can register high confidence on an abrupt or ambiguous call ending; treat completion_confidence as a signal, not a guarantee, and review transcripts for anything time-sensitive
Signals
- GitHub stars
- 104
- Forks
- 528
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
ocmanjos-lead-qualifier- Source
- github.com/calle-ai/awesome-phone-call-agents
github.com/calle-ai/awesome-phone-call-agents
More in Databases & data
Skill · coreyhaines31
More in Databases & datasupabase
Skill · supabase
More in Databases & dataconnect
Skill · composiohq
More in Databases & dataazure-kusto
Skill · microsoft
More in Databases & datarevops
Skill · coreyhaines31
More in Databases & dataagentic-os
Skill · affaan-m
More in Databases & data