Skip to main content

​
sift info

Display project configuration and processing statistics in a formatted table.

​
Usage

sift info

​
Description

The info command provides a comprehensive overview of your current sift-kg project configuration and processing status. It displays information about:
  • Domain configuration (entity types, relation types)
  • LLM model settings
  • Output directory location
  • Processing statistics (documents processed, graph size)
  • Entity resolution status (merge proposals, relation reviews)
  • Narrative generation status
This is useful for:
  • Verifying project configuration
  • Checking processing progress
  • Understanding the current state of your knowledge graph
  • Debugging configuration issues

​
Output

The command displays a formatted table with the following information:
​
Domain
string
The active domain name (e.g., “schema-free”, “general”, “osint”, “academic”)
​
Entity Types
string
Comma-separated list of entity types defined in the domain
​
Relation Types
number
Number of relation types defined in the domain
​
Default Model
string
The configured LLM model (e.g., “openai/gpt-4o-mini”)
​
Output Directory
string
Path to the output directory where all results are stored
​
Documents Processed
number
Number of documents that have been extracted (based on files in output/extractions/)
​
Graph
string
Knowledge graph statistics if built (e.g., “425 entities, 1122 relations”) or “Not built”
​
Merge Proposals
string
Entity merge proposal counts by status (confirmed, draft, rejected) if resolution has been run
​
Relation Review
string
Relation review counts by status (confirmed, draft, rejected) if flagged relations exist
​
Narrative Generated
string
Whether a narrative summary has been generated (“Yes” or “No”)

​
Example Output

sift info

┌─────────────────────────────────────────────────────────────┐
│                    sift-kg Project Info                      │
├──────────────────────┬──────────────────────────────────────┤
│ Metric               │ Value                                 │
├──────────────────────┼──────────────────────────────────────┤
│ Domain               │ schema-free (schema-free)             │
│ Entity Types         │ PERSON, ORGANIZATION, LOCATION...     │
│ Relation Types       │ 15                                    │
│ Default Model        │ openai/gpt-4o-mini                    │
│ Output Directory     │ ./output                              │
│ Documents Processed  │ 12                                    │
│ Graph                │ 425 entities, 1122 relations          │
│ Merge Proposals      │ 5 confirmed, 3 draft, 2 rejected      │
│ Relation Review      │ 8 confirmed, 2 draft, 1 rejected      │
│ Narrative Generated  │ Yes                                   │
└──────────────────────┴──────────────────────────────────────┘

​
Configuration

The command reads configuration from:
  • sift.yaml (if present)
  • Environment variables (.env)
  • Default settings
No command-line options are available. The command always displays information for the current project directory.

​
Use Cases

​
Verify Configuration

Check that your domain and model are configured correctly before starting extraction: 
sift init
sift info  # Verify defaults

​
Check Progress

Monitor processing status during a long-running pipeline: 
sift extract ./documents/
sift info  # See how many documents processed

​
Debug Issues

Understand the current state when troubleshooting: 
sift info
# Check output directory, graph status, domain configuration

​
Review Status

Check resolution and review status before applying merges: 
sift resolve
sift review
sift info  # See how many proposals are confirmed vs draft
sift apply-merges

​
See Also

  • init - Initialize a new project and create sift.yaml
  • build - Build knowledge graph
  • resolve - Find duplicate entities
  • review - Review merge proposals

Build docs developers (and LLMs) love

Get started for freeTalk to us