iron_telemetry 0.4.0

Logging and tracing infrastructure for Iron Cage AI agent management
Documentation

iron_telemetry

Structured logging and tracing for all Iron Cage crates.

Documentation

Installation

[dependencies]
iron_telemetry = { path = "../iron_telemetry" }

Quick Start

use iron_telemetry::{init_logging, log_agent_event, log_pii_detection};

// Initialize logging with default configuration
init_logging()?;

// Log agent lifecycle events
log_agent_event("agent_001", "Agent started processing leads");

// Log PII detection (specialized format)
log_pii_detection("agent_001", "email", "john@example.com");

// Output: [14:32:05] INFO  agent-001 | Agent started processing leads
// Output: [14:32:06] WARN  agent-001 | PII DETECTED: email redacted

Responsibilities: Provides centralized logging abstraction using the tracing crate. Outputs structured, timestamped, colored logs for terminal display and machine-readable JSON for storage. Injects agent context into all log events for traceability.

In Scope:

  • Structured logging via tracing crate
  • Terminal output formatting (timestamp, colors, human-readable)
  • Log levels (DEBUG, INFO, WARN, ERROR, CRIT, OK)
  • Agent context injection (agent_id in all events)
  • Specialized logging functions (PII detections, budget warnings, circuit breaker events)
  • Environment-based log level configuration
  • JSON export for audit compliance

Out of Scope:

  • Log aggregation to external services (Datadog, Splunk, Grafana)
  • Log sampling and filtering (future)
  • Distributed tracing (OpenTelemetry integration)
  • Log rotation and archival (future)
  • Metrics collection (Prometheus, StatsD)
  • Custom log formatters (future)
  • Dashboard log display (see iron_dashboard)

Source Files

File Responsibility
lib.rs Centralized structured logging and tracing for Iron Runtime.

Notes:

  • Entries marked 'TBD' require manual documentation
  • Entries marked '⚠️ ANTI-PATTERN' should be renamed to specific responsibilities

License

Apache-2.0