Applications

Applications Skill

You help users manage their job application tracker. This skill provides conversational access to the application tracker stored in applications.jsonl.

Storage Location

Applications are stored in applications.jsonl in the project root. Each line is a JSON object representing one application.

Data Model

{
  "id": "app-xyz",
  "company": "Company Name",
  "role": "Job Title",
  "status": "applied|interviewing|rejected|offer",
  "date_applied": "2025-01-15",
  "jd_url": "https://...",
  "jd_content": "Full job description...",
  "resume_path": "outputs/company_role/resume.html",
  "company_url": "https://company.com",
  "notes": "Free-form notes",
  "created_at": "...",
  "updated_at": "..."
}

Capabilities

1. List Applications

Read applications.jsonl and present applications in a clear format.

User requests:

• "Show me my applications"
• "List all applications"
• "What applications do I have?"

Response format:

| ID | Company | Role | Status | Date Applied |
|----|---------|------|--------|--------------|
| app-xxx | Company A | Engineer | applied | 2025-01-15 |

2. Show Application Details

Show full details of a specific application.

User requests:

• "Show me app-xxx"
• "Details for the Google application"
• "What's the JD for app-xxx?"

3. Add Application

Create a new application entry. Collect required information interactively.

Required fields:

• Company name
• Role/position

Optional fields:

• JD URL
• JD content (full text)
• Company URL
• Resume path
• Notes

User requests:

• "Add a new application"
• "I applied to Google for a Senior Engineer role"
• "Track my application to [Company]"

Process:

1. If company/role provided, use them
2. Ask for any missing required info
3. Ask for optional info or offer to skip
4. Generate ID (app-XXXXXXXX)
5. Set status to "applied"
6. Set date_applied to today
7. Append to applications.jsonl

4. Update Application

Update status or other fields of an existing application.

Status values:

• applied - Initial state
• interviewing - In interview process
• rejected - Application rejected
• offer - Received offer

User requests:

• "Update app-xxx to interviewing"
• "I got rejected from Google"
• "Mark the Amazon application as offer"
• "Add notes to app-xxx"

5. Remove Application

Remove an application from tracking.

User requests:

• "Remove app-xxx"
• "Delete the Google application"

Always confirm before removing.

6. Natural Language Queries

Answer questions about applications.

User requests:

• "How many applications do I have?"
• "Show pending applications" (status = applied)
• "Which applications are in interview stage?"
• "Applications from last week"
• "Did I apply to Google?"

CLI Commands

Use the bragger CLI for quick operations:

# Interactive mode
bragger add

# Flag mode (quick add)
bragger add --company "Company" --role "Job Title"
bragger add --company "Company" --role "Title" --jd-url "https://..." --notes "Referral"

# Other commands
bragger list             # List all
bragger show <id>        # Show details
bragger update <id>      # Interactive update
bragger update <id> --status "interviewing"  # Flag mode (quick)
bragger remove <id>      # Remove with confirmation

Available flags for add and update:

For add: --company and --role are required when using flags. Without any flags, runs interactively.

For update: All flags are optional. Only provided flags will be updated. Without flags, runs interactively.

Note: --jd-content and --jd-file cannot be used together.

Updating Applications

Update any field using flags:

# Update status
bragger update app-xxx --status "interviewing"

# Update multiple fields
bragger update app-xxx --status "offer" --notes "Accepted the offer!"

# Add JD content later
bragger update app-xxx --jd-file ./jd.txt

Backfilling Old Applications

Use --status and --date to add applications retroactively:

bragger add --company "OldCorp" --role "Engineer" --date "2025-01-15" --status "interviewing"

Adding JD Content

There are three ways to add job description content:

1. From file (recommended for long JDs):

   bragger add --company "Acme" --role "Engineer" --jd-file ./jd.txt
   

2. Inline (for short descriptions):

   bragger add --company "Acme" --role "Engineer" --jd-content "Looking for..."
   

3. Via Claude skill (for fetching from URL): If you only have a URL and want clean extracted content, use the /applications skill:

   "Fetch and store the JD content for the [company] application"

   This leverages Claude's ability to intelligently extract relevant content from job posting pages, filtering out navigation, ads, and other noise.

Implementation Notes

Reading applications.jsonl

Use the Read tool to read the file, then parse each line as JSON.

Writing applications.jsonl

When adding or updating, read all entries, modify as needed, then write back the entire file using the Write tool.

ID Generation

Generate IDs in format: app- + 8 random hex characters (e.g., app-a1b2c3d4)

Example Interactions

User: "I just applied to Stripe for a Senior Backend Engineer role"

Response:

1. Create new application with company="Stripe", role="Senior Backend Engineer"
2. Set status="applied", date_applied=today
3. Ask: "Would you like to add the job description or any notes?"
4. Save to applications.jsonl
5. Confirm: "Added application app-xxx for Senior Backend Engineer at Stripe"

User: "Show me my pending applications"

Response:

1. Read applications.jsonl
2. Filter where status="applied"
3. Display in table format
4. Include count: "You have X pending applications"

User: "I got an interview with Stripe!"

Response:

1. Find Stripe application
2. Update status to "interviewing"
3. Confirm: "Updated Stripe application to interviewing status"