AI Sleep Coach Agent Skill — AI Pass API
Overview
AI Pass "AI Sleep Coach" skill (sleep-coach) analyzes a user's sleep patterns and provides evidence-based, personalized sleep coaching advice. The skill ingests sleep logs and pattern descriptions (sleep duration, awakenings, bedtime/wake time variability, naps, caffeine/alcohol/timing), then returns actionable recommendations, sleep-hygiene tips, and suggested bedtime routines tailored to the individual.
Use cases for AI agents:
- Personalized sleep assessments for digital health assistants or wellness apps.
- Integrating automated sleep-coaching into telehealth or behavioral health workflows.
- Generating weekly sleep improvement plans and progress summaries from wearable/exported sleep data.
- Triage guidance and sleep hygiene education for coaches, therapists, or fitness platforms (not a replacement for medical advice).
Skill purpose: Analyze user sleep patterns and generate evidence-based, personalized sleep coaching advice.
API base URL: https://aipass.one/apikey/v1
Developer dashboard: https://aipass.one/panel/developer.html
Skill File (YAML)
name: sleep-coach
description: Analyzes your sleep patterns and provides personalized sleep improvement advice
version: "1.0"
auth:
type: bearer
token_env: AIPASS_API_KEY
endpoints:
- name: generate
method: POST
url: https://aipass.one/apikey/v1/chat/completions
headers:
Authorization: "Bearer $AIPASS_API_KEY"
Content-Type: application/json
(ALWAYS use $AIPASS_API_KEY — never a real key)
Example Usage
Curl example
curl -X POST "https://aipass.one/apikey/v1/chat/completions" \
-H "Authorization: Bearer $AIPASS_API_KEY" \
-H "Content-Type: application/json" \
-H "X-Client-Id: YOUR_CLIENT_ID" \
-d '{
"model": "aipass-gpt-4o",
"messages": [
{"role": "system", "content": "You are a sleep coach. Provide evidence-based, practical sleep improvement advice."},
{"role": "user", "content": "Here is my sleep log for the past week: Mon 6.0h, Tue 5.5h, Wed 7.0h, Thu 4.5h (woke up 2 times), Fri 6.5h, Sat 8.5h, Sun 7.5h. Bedtime varies between 11pm and 2am. I drink coffee at 3pm most days."}
],
"temperature": 0.2,
"max_tokens": 600
}'
Python example
import os
import requests
API_URL = "https://aipass.one/apikey/v1/chat/completions"
API_KEY = os.environ.get("AIPASS_API_KEY") # ALWAYS use the environment variable
HEADERS = {
"Authorization": f"Bearer {API_KEY}",
"Content-Type": "application/json",
"X-Client-Id": "YOUR_CLIENT_ID"
}
payload = {
"model": "aipass-gpt-4o",
"messages": [
{"role": "system", "content": "You are a sleep coach. Give evidence-based, personalized advice and practical next steps."},
{"role": "user", "content": (
"Sleep log: Mon 6.0h, Tue 5.5h, Wed 7.0h, Thu 4.5h (2 awakenings), "
"Fri 6.5h, Sat 8.5h, Sun 7.5h. Bedtime varies 11pm-2am. Coffee at 3pm daily. "
"Goal: feel more rested and reduce mid-night awakenings."
)}
],
"temperature": 0.2,
"max_tokens": 600
}
resp = requests.post(API_URL, headers=HEADERS, json=payload)
resp.raise_for_status()
result = resp.json()
# The model response will typically be in result["choices"][0]["message"]["content"]
print(result)
Notes:
- Include "X-Client-Id: YOUR_CLIENT_ID" if you need to pass a client identifier.
- Adjust model, temperature, and max_tokens to tune style and length of the coaching response.
- Always sanitize and limit any PHI/personal data per your privacy policy and applicable regulations.
Available Models
Relevant AI Pass models you can use with this skill (choose based on quality/latency/cost):
- aipass-gpt-4o — best overall quality for nuanced, evidence-based coaching and multi-step plans.
- aipass-gpt-4-mini — lower-latency, cost-effective option for conversational coaching and summaries.
- aipass-gpt-3.5-turbo — fast and economical for basic recommendations and short tips.
- aipass-embed-1 — use for embedding sleep logs or user profiles to build personalized retrieval-augmented coaching (paired with a vector DB).
Get Your API Key
Get an API key from the developer dashboard: https://aipass.one/panel/developer.html → API Keys section.
Remember: NEVER commit or share real API keys in code repositories. Use $AIPASS_API_KEY in examples and store keys in secure environment variables for production.
Related Apps
Explore the app page: https://aipass.one/apps/sleep-coach
Important reminders
- The AI Sleep Coach provides coaching and educational material, not medical diagnoses. Advise users to seek professional medical care for serious or persistent sleep disorders.
- ALWAYS use $AIPASS_API_KEY (environment variable) and YOUR_CLIENT_ID placeholders in examples and configs — never embed real secrets.
Skill File
name: ai_pass_sleep_coach
title: AI Pass — Sleep Coach
version: 1.0
frameworks:
- OpenClaw
- generic
description: >
Analyzes your sleep patterns and provides personalized sleep improvement advice.
Use this skill to send sleep logs, symptoms, and questions to the AI Pass chat
completions API and return actionable, evidence-based sleep coaching guidance.
auth:
type: bearer
env_var: AIPASS_API_KEY
header_name: Authorization
header_value_prefix: "Bearer "
api:
base_url: https://aipass.one/apikey/v1
timeout_seconds: 30
endpoints:
- id: chat_completions
name: Create Chat Completion
description: Call AI Pass chat completions to generate personalized sleep-coach responses.
method: POST
path: /chat/completions
headers:
Content-Type: application/json
Authorization: "Bearer ${AIPASS_API_KEY}"
request_schema:
type: object
description: Request body follows an OpenAI-like chat completions schema.
properties:
model:
type: string
description: Model name to use (choose a model supported by AI Pass)
default: gpt-4o-mini
messages:
type: array
description: Array of message objects (role, content) in chat format.
items:
type: object
properties:
role:
type: string
enum: [system, user, assistant]
content:
type: string
temperature:
type: number
minimum: 0
maximum: 2
default: 0.7
max_tokens:
type: integer
default: 800
required:
- model
- messages
response_schema:
type: object
description: >
The API returns a chat completion response. The assistant reply text is
typically found at choices[0].message.content.
properties:
raw:
type: object
description: Full raw response from the AI Pass API.
assistant_text:
type: string
description: Extracted assistant reply (choices[0].message.content)
example: "Here are some personalized recommendations to improve your sleep..."
error_handling:
- status: 401
message: "Invalid or missing API key. Ensure AIPASS_API_KEY is set."
- status: 429
message: "Rate limited. Consider backing off and retrying."
- status: 500
message: "Server error from AI Pass. Retry with exponential backoff."
examples:
- name: sleep_log_analysis
description: Send sleep diary and ask for tailored improvement plan
request:
model: gpt-4o-mini
messages:
- role: system
content: >-
You are a Sleep Coach AI. Analyze sleep logs, identify patterns, and provide
concise, evidence-based, actionable advice including sleep hygiene,
schedule changes, strategies for nighttime awakenings, and CBT-I techniques
when appropriate.
- role: user
content: |
I usually go to bed at 1:00 AM and wake at 7:00 AM (about 6 hours). I wake
around 3:00 AM and have trouble falling back asleep. I drink coffee at 3 PM,
use my phone in bed, and take a 30-minute nap most afternoons. I feel tired
daytime and want to improve sleep. What should I change?
expected_response:
assistant_text: >-
A personalized plan including: move caffeine earlier (before 2pm), establish a
consistent wind-down routine removing screens 30–60 minutes before bed, adjust
sleep window gradually to increase total sleep, strategies for brief awakenings
(get out of bed if unable to fall back asleep within ~20 minutes), limit naps,
improve sleep environment, and consider short-term sleep restriction or CBT-I
if insomnia persists. Include measurable steps and timeline.
notes:
- Ensure the environment variable AIPASS_API_KEY is set in the agent runtime.
- This skill posts to /chat/completions and expects OpenAI-like chat messages.
- Keep user messages concise and provide sleep logs, patterns, durations, and relevant behaviors for best results.