confirmcall-day-board
Confirm a full day's existing appointments by phone with CALL-E. Plans one disclosed confirmation call per booking, captures yes/no/reschedule as structured JSON, and leaves calendar writes to a human. Fixture/dry-run by default.
- License 许可
- MIT
- 最近更新
- 2026-10-08
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- confirmcall-day-board
- license
- MIT
- description
- Confirm a full day's existing appointments by phone with CALL-E. Plans one disclosed confirmation call per booking, captures yes/no/reschedule as structured JSON, and leaves calendar writes to a human. Fixture/dry-run by default.
02 资源文件 4
- 资产文件assets/sample-day-board.json1.4 KB
- 参考文档references/call-script.md614 B
- 参考文档references/examples.md605 B
- 参考文档references/safety.md610 B
03 SKILL.md 正文
ConfirmCall Day Board
Use this skill when a clinic, tutor, or service desk has a day list of existing appointments and explicit authority to place disclosed confirmation calls. The skill turns each booking into one CALL-E confirmation task, runs them (or returns fixtures), and writes yes / no / reschedule results back to a board for a human.
This skill does not book, move, or cancel calendar events. A human reviews results before any diary change.
When To Use
- Morning confirmation pass over today's booked appointments
- Capture declines and reschedule requests into structured results
- Preview every CALL-E script before any live dial
- Keep calendar authority with a human coordinator
When Not To Use
- First-contact sales, lead qualification, or unsolicited outreach
- Medical advice, legal, financial, emergency, collections, or political calls
- Calling numbers the coordinator did not authorize
- Hidden retries, recurring campaigns, or automatic calendar writes
- Collecting payment, ID numbers, or health details on the call
Required Inputs (per appointment)
request_id: stable local identifierbusiness_display_name: disclosed on the callrecipient_first_nameto_phone_e164: E.164, authorizedappointment.service,appointment.starts_at(ISO-8601 with offset),appointment.locationtimezone: IANAconsent: must be trueauthorized_reason: why this specific call is allowed
Optional: reschedule_windows (max 6 ISO-8601 times the desk can honor).
Safety Model
- Fixture / dry-run by default — no network, no CALL-E dial.
- Live mode only when
CALLE_API_KEY(orCALL_E_API_KEY) is set and the operator explicitly runs live. - Refuse if
consentis not true or the number is not on the day's authorized list. - Mask phone numbers in logs (
+1***0100). - One call plan per appointment — no silent retries.
- Calendar remains human-owned.
Dry-Run (no CALL-E credentials)
From the repository root:
cd apps/typescript/confirmcall-day-board
npm install
npm run fixture
This loads the Thursday sample board, prints planned CALL-E scripts, and writes fixture outcomes (Maya confirmed, Jordan reschedule, Eli declined).
Live Path
- Create a CALL-E API key in the CALL-E dashboard.
- Export
CALLE_API_KEY. - Only call numbers you are authorized to contact.
- Run
npm run liveonly after reviewingnpm run previewscripts.
Prefer the runnable app under apps/typescript/confirmcall-day-board for the full day-board UX. Use this skill when an agent host should orchestrate the same workflow via CALL-E SDK/API/MCP.
Result Shape
Each appointment returns fail-closed structured JSON:
{
"request_id": "apt-maya-1",
"disposition": "confirmed",
"notes": "Recipient confirmed the booked slot",
"requested_time": null
}
disposition is one of confirmed, declined, reschedule_requested, no_answer, failed.
References
references/call-script.md— disclosed confirmation script templateassets/sample-day-board.json— fictional Thursday board
04 同仓库更多 Skills
xitter
Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.
writing-plans
Use when you have a spec or requirements for a multi-step task. Creates comprehensive implementation plans with bite-sized tasks, exact file paths, and complete code examples.
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.
weights-and-biases
Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
web-pentest
Authorized web pentest: recon, proof-based exploits, report.