Dancer - Self-Modifying OCaml Services#

Dancer is an ambitious OCaml library that enables long-running services to automatically improve themselves through AI-assisted code generation, with comprehensive logging, pattern detection, and safe deployment mechanisms.

Components#

dancer-logs#

Comprehensive structured logging with SQLite backend. Features:

• Full structured logging with no truncation
• SQLite storage with FTS5 full-text search
• Pattern detection and normalization
• Eio fiber and domain tracking
• Rich context capture (sessions, requests, traces, performance metrics)
• OCaml Logs reporter integration
• Colorful CLI for log analysis

Installation#

# Install dependencies
make deps

# Build the project
make build

# Run the CLI
make run-cli

Usage#

(* Initialize logging *)
let db = Dancer_logs.init ~path:"app.db" () in
let reporter = Dancer_logs.reporter db in
Logs.set_reporter reporter;

(* Log with context *)
let ctx = Dancer_logs.Context.(
  empty_context
  |> with_session "session-123"
  |> with_request "req-456"
) in

Dancer_logs.log db ~level:Logs.Error ~source:"MyApp" 
  ~message:"Connection failed" ~context:ctx ()

CLI Usage#

# View recent errors
dancer-logs errors

# Search logs
dancer-logs search "connection refused"

# View patterns
dancer-logs patterns --min-count 10

# Export for Claude
dancer-logs export -f claude -o context.txt

Architecture#

See CLAUDE.md for detailed architecture and design decisions.

License#

MIT