resume - Resume Interrupted Jobs
Resume interrupted scraping jobs from saved checkpoints and manage job history.
Basic Usage
skill-seekers resume [OPTIONS] [JOB_ID]
Quick Examples
# List all resumable jobs
skill-seekers resume --list
# Resume specific job by ID
skill-seekers resume abc123def456
# Clean up old job files
skill-seekers resume --clean
# View job details
skill-seekers resume --list --verbose
Options
Actions
--list- List all resumable jobs with progress details--clean- Clean up old job files (respects config settings)JOB_ID- Resume specific job from checkpoint
Display Options
--verbose- Show detailed job information--format FORMAT- Output format: table, json, simple (default: table)
Resume Feature
When Jobs are Saved
Skill Seekers automatically saves progress for resumable operations:
Auto-saved operations:
- Documentation scraping (checkpoint every N pages)
- GitHub repository analysis (checkpoint after each phase)
- PDF extraction (checkpoint after each file)
- Multi-source unified scraping (checkpoint per source)
Save interval:
- Default: 60 seconds
- Configurable:
skill-seekers config→ Resume Settings
Progress Storage
Location:
~/.local/share/skill-seekers/progress/<job-id>.json
Job File Structure:
{
"job_id": "abc123def456",
"command": "skill-seekers github --repo facebook/react",
"started_at": "2026-01-18T10:30:00Z",
"last_updated": "2026-01-18T10:45:00Z",
"progress": {
"phase": "Code Analysis",
"files_processed": 1234,
"files_total": 2000,
"percent_complete": 61.7
},
"checkpoints": {
"scraping_complete": true,
"analysis_phase_1": true,
"analysis_phase_2": false
},
"metadata": {
"repo": "facebook/react",
"output_dir": "output/react"
}
}
Listing Jobs
Basic List
skill-seekers resume --list
Output:
Resumable Jobs (3 found)
─────────────────────────────────────────────────────────────────
JOB ID: abc123def456
Started: 2026-01-18 10:30:00
Command: skill-seekers github --repo facebook/react
Progress: Code Analysis (61.7% - 1234/2000 files)
Last Updated: 2 minutes ago
JOB ID: def456ghi789
Started: 2026-01-17 15:20:00
Command: skill-seekers scrape https://docs.astro.build
Progress: Scraping (450/500 pages)
Last Updated: 1 day ago
JOB ID: ghi789jkl012
Started: 2026-01-16 09:00:00
Command: skill-seekers unified --config configs/godot_full.json
Progress: Source 2 of 3 (GitHub Analysis)
Last Updated: 2 days ago
Verbose List
skill-seekers resume --list --verbose
Shows additional details:
- Full command with arguments
- All checkpoint statuses
- Estimated time remaining
- Output directory path
- Error logs (if any)
JSON Format
skill-seekers resume --list --format json
Use cases:
- Scripting and automation
- Integration with other tools
- Parsing in CI/CD pipelines
Resuming Jobs
Resume by Job ID
skill-seekers resume abc123def456
What happens:
- Loads job metadata from progress file
- Validates checkpoint integrity
- Resumes from last successful checkpoint
- Continues operation as normal
- Updates progress file automatically
Resume Latest Job
# List and resume the most recent job
skill-seekers resume --list
skill-seekers resume $(skill-seekers resume --list --format json | jq -r '.[0].job_id')
Resume After Rate Limit
If a GitHub scraping job hit rate limits:
# Check job status
skill-seekers resume --list
# Wait for rate limit reset or switch profile
skill-seekers config --github # Add another profile
# Resume with different profile
skill-seekers resume abc123def456 --profile work
Clean Up
Auto-Cleanup
Configured in skill-seekers config:
Default settings:
- Auto-cleanup age: 7 days
- Jobs older than 7 days are automatically removed on next run
Manual Cleanup
# Remove all jobs older than configured age
skill-seekers resume --clean
Output:
Cleaning up old job files...
✓ Removed abc123def456 (started 10 days ago)
✓ Removed def456ghi789 (started 15 days ago)
Kept ghi789jkl012 (started 2 days ago)
Cleaned up 2 job files.
Force Cleanup All
# Remove all job files regardless of age
skill-seekers resume --clean --all
Warning: This deletes ALL resumable jobs, including recent ones.
Resume Scenarios
Scenario 1: Network Interruption
Problem: Internet disconnected during documentation scraping
# Original command
skill-seekers scrape https://docs.django.com --max-pages 1000
# ... network interruption at page 450 ...
# Resume after network restored
skill-seekers resume --list
skill-seekers resume abc123def456
Result: Resumes from page 450, skips already-scraped pages
Scenario 2: Rate Limit Hit
Problem: GitHub API rate limit exceeded during repository analysis
# Original command
skill-seekers github --repo microsoft/vscode
# ... rate limit hit after 30 minutes ...
# Option 1: Wait and resume
# (waits for rate limit reset)
skill-seekers resume abc123def456
# Option 2: Switch profile and resume
skill-seekers config --github # Add new profile
skill-seekers resume abc123def456 --profile work
Result: Continues analysis from last checkpoint
Scenario 3: System Crash
Problem: Computer crashed during unified scraping
# Original command
skill-seekers unified --config configs/godot_full.json
# ... system crash ...
# After reboot, list jobs
skill-seekers resume --list
# Resume from checkpoint
skill-seekers resume abc123def456
Result: Resumes from last auto-save (default: 60 seconds)
Scenario 4: Manual Cancellation
Problem: Accidentally canceled long-running job
# Press Ctrl+C during execution
# Later, resume it
skill-seekers resume --list
skill-seekers resume abc123def456
Result: Picks up where it left off
Progress Tracking
Understanding Progress
Phase indicators:
Initialization- Setting up scraping environmentScraping- Fetching pages/filesCode Analysis- AST parsing (GitHub repos only)C3.x Analysis- Design patterns, test extraction (if enabled)Enhancement- AI enhancement phasePackaging- Finalizing skill package
Progress metrics:
- Files/pages processed vs total
- Percent complete
- Current phase
- Estimated time (verbose mode)
Real-Time Updates
While running, resumed jobs show progress:
Resuming job abc123def456...
Loaded checkpoint: Code Analysis (phase 1 complete)
Processing files: [████████░░░░░░░░] 1500/2000 (75%)
Current file: src/components/Button.tsx
Elapsed: 45m 23s | Remaining: ~15m 12s
Resume Configuration
Configure resume behavior in skill-seekers config:
Auto-Save Interval
Default: 60 seconds
skill-seekers config
# → Select "4. Resume Settings"
# → Select "Auto-save interval"
# → Enter new value (30-300 seconds recommended)
Trade-offs:
- Lower interval (30s) - More frequent saves, better resume granularity, slight performance impact
- Higher interval (120s) - Less frequent saves, may lose more progress on crash
Auto-Cleanup Age
Default: 7 days
skill-seekers config
# → Select "4. Resume Settings"
# → Select "Auto-cleanup age"
# → Enter new value (1-30 days recommended)
Recommendations:
- Development: 7 days (default)
- CI/CD: 1 day (clean up quickly)
- Production: 14 days (keep longer history)
Limitations
Not Resumable
These operations cannot be resumed:
enhancecommand (API-based enhancement)packagecommand (fast operation)uploadcommand (quick upload)
Partial Resume
These operations resume with limitations:
- AI enhancement (LOCAL mode) - Resumes per-file, may re-enhance some files
- Parallel scraping - May re-scrape pages from interrupted batch
Integration with Other Commands
GitHub Command
# Start GitHub scraping
skill-seekers github --repo facebook/react
# If interrupted, resume it
skill-seekers resume <job-id>
# Resume with different options
skill-seekers resume <job-id> --profile work --non-interactive
Unified Command
# Start multi-source scraping
skill-seekers unified --config configs/framework.json
# If interrupted during source 2, resume
skill-seekers resume <job-id>
# Continue from checkpoint: source 2 analysis
Troubleshooting
Job Not Found
# Verify job exists
skill-seekers resume --list
# Check progress file location
ls -la ~/.local/share/skill-seekers/progress/
# Recreate progress directory if missing
mkdir -p ~/.local/share/skill-seekers/progress/
Corrupted Checkpoint
# If resume fails with "corrupted checkpoint":
# Option 1: Start fresh
skill-seekers github --repo owner/repo --output output/repo_new
# Option 2: Delete corrupted checkpoint
rm ~/.local/share/skill-seekers/progress/<job-id>.json
Permission Error
# Fix permissions on progress directory
chmod 700 ~/.local/share/skill-seekers/progress/
chmod 600 ~/.local/share/skill-seekers/progress/*.json
Best Practices
- Check job list regularly - Use
--listto see pending jobs - Clean up old jobs - Run
--cleanmonthly to free disk space - Use verbose mode for debugging -
--list --verboseshows full details - Resume promptly - Jobs older than cleanup age are auto-removed
- Verify checkpoints - Resume command validates integrity before continuing
- Use shorter auto-save intervals - For unstable networks or systems
- Keep job history - For production, increase cleanup age to 14-30 days
See Also
- config command - Configure resume settings
- github command - GitHub repository scraping
- Rate Limit Management - Handling rate limits with resume