Shop Voice Checkin

What it does

When your agent uses it

• Business, Finance & HR work in your project

Example prompts

• “/shop-voice-checkin”

Requirements

• Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Read references/safety.md and confirm recipient consent.
2. 2Choose call type: inventory (morning) or sales (evening).
3. 3Build task text from references/call-scripts-pidgin.md, references/call-scripts-english.md, or references/call-scripts-hindi-english.md.
4. 4Attach the matching result schema
5. 5Preview first — inspect the planned task and schema without placing a call.
6. 6Live call — only with explicit user approval and --execute-style confirmation in the runnable app.
7. 7Pass structured results to the shop ledger app or host workflow.

What it can do on your machine

Read from SKILL.md and the folder at commit 38d4118. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/, which the agent can run.

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Shop Voice Checkin loads about 2.4k tokens when it runs, and up to ~14k if it reads all its reference files. Until then it costs about 69 tokens; SKILL.md has 708 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~69

When it runs · the whole SKILL.md, loaded when a task matches

~2.4k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~14k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from CALLE-AI/awesome-phone-call-agents at commit 38d4118, republished under its MIT licence (© CALLE-AI). 708 words, ~2,427 tokens.

name

shop-voice-checkin

description

Place consent-based outbound CALL-E phone check-ins with informal African and Indian retailers to capture morning inventory and evening sales through natural voice conversation, then return structured shop data for a voice-first business manager workflow.

license

MIT

consent check -> build task + schema -> preview -> live call -> structured result -> shop ledger

Call the shop owner for a short morning inventory check-in. Match language_style:
Pidgin for pidgin-english, plain English for english, Hindi-English mix for
hindi-english. Ask about: {{products_to_ask}}. For each product, capture
approximate quantity and unit. Ask what is running low. Also ask once whether
they added any new goods not on that list; if yes, capture name, quantity, and
unit so the ledger can add them. Disclose you are an AI assistant for their shop
manager service. Keep the call under {{max_minutes}} minutes. Do not give
financial advice.

Call the shop owner for a short evening sales recap. Ask roughly how much they
sold today, what sold best, and whether they bought stock for the shop today.
Disclose you are an AI assistant for their shop manager service. Keep the call
under {{max_minutes}} minutes. Do not give financial advice.

Call the consenting shop owner. Tell them the following is running low:
{{low_stock_items}}. Ask if they want to place a restock order. If yes, ask quantity
needed per item, then ask which vendor to use. If it is a vendor already on file, do
not ask what they sell or their phone number again. If it is a new vendor, explicitly
ask two more questions before ending the call: what the vendor sells, and the
vendor's phone number — do not skip the phone number just because the owner did not
offer it; ask for it directly. If the owner truly does not have it, say the vendor
cannot be called yet without it. Never invent a vendor phone number. If no, end
politely; do not place any vendor call. Disclose you are an AI assistant. Keep the
call under {{max_minutes}} minutes. Do not give financial advice or discuss loans.

Call {{vendor_display_name}} on behalf of {{shop_id}}. Disclose you are an AI
assistant calling on their behalf. State you are calling to place an order for
{{order_items}}. Ask if they are available, the price if they wish to share
it, and an estimated delivery time. Do not discuss payment, bank details, or
loans.

Call the consenting shop owner with a short update on their restock order.
Disclose you are an AI assistant. Tell them exactly the known outcome from the
vendor call — placed, unavailable, or delayed — plus the ETA and amount if
known. Do not leave the outcome for the model to guess; a task with no known
outcome to report should not be created. Keep under 2 minutes. Do not give
financial advice.

Call the consenting new shop owner. Disclose you are an AI assistant. Explain
you will collect basic shop details to set up their account: shop name, phone
number (read back to confirm), region, and language. Ask for 3 to 5 staple
products they sell and, optionally, one preferred supplier's name. Ask for
consent to store this information and to make future check-in calls. Do not
infer region, locale, or currency — ask explicitly. Keep under {{max_minutes}}
minutes.

shopvoice-{shop_id}-inventory-{YYYY-MM-DD}
shopvoice-{shop_id}-sales-{YYYY-MM-DD}
shopvoice-{shop_id}-reorder_offer-{YYYY-MM-DD}
shopvoice-{shop_id}-vendor_order-{request_id}
shopvoice-{shop_id}-order_status-{request_id}
shopvoice-{shop_id}-onboarding-{YYYY-MM-DD}

© CALLE-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 20 other files (scripts, references, assets) in skills/shop-voice-checkin of CALLE-AI/awesome-phone-call-agents.

• SKILL.md
• assets/example-request-inr.template.json
• assets/example-request.template.json
• assets/sample-shop-profile-inr.json
• assets/sample-shop-profile.json
• references/call-scripts-english.md
• references/call-scripts-hindi-english.md
• references/call-scripts-pidgin.md
• references/demo-mode.md
• references/examples.md
• references/result-schema-inventory.json
• references/result-schema-onboarding.json
• references/result-schema-order-status.json
• references/result-schema-payment-consent.json
• references/result-schema-reorder-offer.json
• references/result-schema-sales.json
• references/result-schema-vendor-order.json
• references/safety.md
• references/scheduling.md
• … and 2 more

Open the folder on GitHubat commit 38d4118