AI Journal
Click to zoom
AI Journal
Your personal AI-powered journaling companion that analyzes entries, tracks mood patterns, and provides compassionate weekly insights.
Meet your new journaling companion — an AI-powered personal journal that goes beyond simple note-taking. AI Journal understands your emotions, tracks your mental well-being, and provides supportive, non-judgmental feedback like having a compassionate therapist in your pocket.
Why AI Journal?
- Emotional Intelligence: Advanced sentiment analysis detects mood, energy levels, and stress indicators
- Theme Tracking: Automatically identifies recurring topics, concerns, and gratitudes
- Supportive Responses: Warm, validating feedback that encourages self-reflection
- Weekly Insights: Aggregated mood trends, patterns, and personalized encouragement
- Privacy First: Your entries stay on your device, processed locally
Whether you're practicing mindfulness, working through challenges, or simply want to build a journaling habit, AI Journal makes the experience more meaningful and insightful.
Workflow Visualization
Requirements
- OpenAI API key (for GPT-4o)
Key Files
yaml
version: "1.0"
workflows:
- id: save-entry
path: ./habits/save-entry.yaml
enabled: true
- id: get-entries
path: ./habits/get-entries.yaml
enabled: true
- id: weekly-insights
path: ./habits/weekly-insights.yaml
enabled: true
- id: get-entry
path: ./habits/get-entry.yaml
enabled: true
server:
port: 13000
host: "0.0.0.0"
frontend: ./frontend
openapi: true
logging:
level: info
outputs: [console]
format: text
colorize: trueyaml
id: get-entries
name: Get Journal Entries
description: Get journal entries for a user
input:
- id: userId
type: string
required: false
- id: limit
type: number
required: false
- id: startDate
type: string
required: false
- id: endDate
type: string
required: false
nodes:
- id: query-entries
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-database"
resource: query
operation: query
params:
collection: "journal_entries"
limit: "{{habits.input.limit}}"
sort: '{"createdAt": -1}'
edges: []
output:
entries: "{{query-entries.results}}"
count: "{{query-entries.count}}"yaml
id: get-entry
name: Get Journal Entry
description: Get a specific entry by ID
input:
- id: id
type: string
required: true
nodes:
- id: fetch-entry
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-database"
resource: findOne
operation: findOne
params:
collection: "journal_entries"
filter: '{"_id": "{{habits.input.id}}"}'
edges: []
output:
entry: "{{fetch-entry.document}}"
found: "{{fetch-entry.found}}"yaml
id: save-entry
name: Save Journal Entry
description: Save a journal entry with AI analysis for mood, themes, and supportive feedback
input:
- id: content
type: string
required: true
description: The journal entry content
- id: userId
type: string
required: false
description: User identifier
nodes:
# Analyze sentiment/mood
- id: analyze-mood
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-intersect"
resource: ask_chatgpt
operation: ask_chatgpt
credentials:
intersect:
host: "{{habits.env.HABITS_INTERSECT_HOST}}"
apiKey: "{{habits.env.HABITS_INTERSECT_API_KEY}}"
params:
model: gpt-4o
temperature: 0.3
maxTokens: 300
systemMessage: You are a compassionate therapist analyzing journal entries. Be supportive and insightful.
prompt: |
Analyze the emotional tone of this journal entry:
{{habits.input.content}}
Return JSON with:
{
"moodScore": <1-10 scale, 1=very negative, 10=very positive>,
"primaryEmotion": "main emotion expressed",
"secondaryEmotions": ["other emotions detected"],
"energyLevel": "low/medium/high",
"stressIndicators": true/false
}
Return ONLY the JSON.
# Extract themes
- id: extract-themes
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-intersect"
resource: ask_chatgpt
operation: ask_chatgpt
credentials:
intersect:
host: "{{habits.env.HABITS_INTERSECT_HOST}}"
apiKey: "{{habits.env.HABITS_INTERSECT_API_KEY}}"
params:
model: gpt-4o
temperature: 0.3
maxTokens: 300
prompt: |
Extract key themes and topics from this journal entry:
{{habits.input.content}}
Return JSON:
{
"themes": ["work", "relationships", "health", etc.],
"topics": ["specific things mentioned"],
"people": ["names of people mentioned"],
"gratitudes": ["things they're grateful for, if any"],
"concerns": ["worries or problems mentioned"]
}
Return ONLY the JSON.
# Generate supportive response
- id: generate-response
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-intersect"
resource: ask_chatgpt
operation: ask_chatgpt
credentials:
intersect:
host: "{{habits.env.HABITS_INTERSECT_HOST}}"
apiKey: "{{habits.env.HABITS_INTERSECT_API_KEY}}"
params:
model: gpt-4o
temperature: 0.7
maxTokens: 500
systemMessage: |
You are a compassionate, wise journaling companion. Provide supportive,
thoughtful responses that validate feelings, offer gentle insights, and
encourage self-reflection. Never be preachy or give unsolicited advice.
Be warm, human, and understanding.
prompt: |
The user wrote this journal entry:
{{habits.input.content}}
Their mood analysis: {{analyze-mood}}
Themes identified: {{extract-themes}}
Write a supportive, personalized response (2-3 paragraphs):
1. Acknowledge and validate their feelings
2. Reflect back any insights you notice
3. End with an encouraging thought or gentle question for reflection
Be genuine and warm, not clinical.
# Generate reflection prompts
- id: reflection-prompts
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-intersect"
resource: ask_chatgpt
operation: ask_chatgpt
credentials:
intersect:
host: "{{habits.env.HABITS_INTERSECT_HOST}}"
apiKey: "{{habits.env.HABITS_INTERSECT_API_KEY}}"
params:
model: gpt-4o
temperature: 0.8
maxTokens: 200
prompt: |
Based on this journal entry:
{{habits.input.content}}
Suggest 3 thoughtful reflection questions for the user to consider.
Make them specific to what they wrote about.
Return as JSON: {"prompts": ["question 1", "question 2", "question 3"]}
# Save entry
- id: save-entry
type: bits
data:
framework: bits
source: npm
module: "@ha-bits/bit-database"
resource: insert
operation: insert
params:
collection: "journal_entries"
document:
userId: "{{habits.input.userId}}"
content: "{{habits.input.content}}"
mood: "{{analyze-mood}}"
themes: "{{extract-themes}}"
response: "{{generate-response}}"
reflectionPrompts: "{{reflection-prompts}}"
createdAt: "{{habits.now}}"
edges:
- source: analyze-mood
target: generate-response
- source: extract-themes
target: generate-response
- source: generate-response
target: reflection-prompts
- source: reflection-prompts
target: save-entry
output:
entryId: "{{save-entry.id}}"
mood: "{{analyze-mood}}"
themes: "{{extract-themes}}"
response: "{{generate-response}}"
reflectionPrompts: "{{reflection-prompts}}"Quick Start
Run directly using Cortex package, recommended for production runs, does not inlcude base or extra depdencies.
# First, download the example files
npx @ha-bits/cortex@latest server --config ./ai-journal/stack.yaml