Overview

/evo-stats is a production command in the Framework category of the Prismatic Platform. It provides comprehensive evolution statistics and metrics reporting, giving operators deep visibility into the platform's autonomous evolution trajectory across all 18 generations. The command aggregates data from every evolution cycle, quality gate checkpoint, and fitness evaluation to produce actionable intelligence about the platform's evolutionary health.

The Prismatic Platform operates on a biological evolution metaphor where the codebase, agents, patterns, and quality standards undergo continuous generational improvement. Each generation represents a measurable fitness improvement across multiple dimensions: code quality, test coverage, agent capability, pattern diversity, and architectural coherence. The /evo-stats command is the primary observability tool for this evolutionary process, transforming raw evolution data into structured reports that inform strategic decisions.

Understanding evolution statistics is essential for maintaining the platform's upward trajectory. Without visibility into fitness trends, mutation rates, and selection pressures, the evolution system risks stagnation or regression. The /evo-stats command surfaces early warning signals when fitness gains plateau, when specific quality domains lag behind others, or when evolution cycles are not producing meaningful improvements.

This command operates under the L2+ authority level and is executed by the evolution-orchestrator agent. It is part of the platform's 216-command slash command registry, built on the AIAD (Autonomous Intelligence Agent Design) standard.

Architecture

The evolution statistics system aggregates data from multiple platform subsystems into a unified reporting pipeline:

Evolution Engine --> Fitness Evaluator --> Metrics Aggregator --> Report Generator
      |                    |                      |                     |
 Generation DB       Quality Scores          Time Series           Visualization
      |                    |                      |                     |
 Gen 1..18 Data     13 Quality Domains      Trend Analysis        Charts/Tables
      \                    |                      /                     |
       --> Statistics Synthesizer --> Dashboard --> Output Formatter
                    |
              Anomaly Detector

Fitness Evaluator: Computes multi-dimensional fitness scores across the platform's 13 quality domains (Dialyzer, Credo, Compilation, DateTime Precision, Guard Functions, @impl Coverage, Memory Safety, Performance, Regression Prevention, Timing Patterns, TODO Management, Typespec Coverage, Unsafe Map Access). Each domain contributes a normalized score to the overall platform fitness.

Metrics Aggregator: Collects evolution metrics from all subsystems including the SEADF framework, quality gates, mycelial network, and Quality DNA persistence layer. Metrics are stored as time series for trend analysis.

Anomaly Detector: Monitors evolution metrics for anomalous patterns -- fitness regressions, stalled evolution cycles, quality domain divergence, or unusual mutation rates. Anomalies trigger alerts that surface in the report.

Report Generator: Produces structured reports in multiple formats (text, JSON, table, markdown) with configurable detail levels from executive summary to full diagnostic drill-down.

Usage

Quick Status Check

# Display current evolution status summary
/evo-stats

# Show current generation fitness
/evo-stats --current

# Brief one-line status
/evo-stats --brief

Detailed Analysis

# Full evolution history across all generations
/evo-stats --full-history

# Quality domain breakdown for current generation
/evo-stats --domains --verbose

# Fitness trend analysis over last 5 generations
/evo-stats --trend --generations=5

# Compare two specific generations
/evo-stats --compare --gen1=15 --gen2=18

Reporting

# Generate JSON report for CI/CD integration
/evo-stats --format=json --output=evolution-report.json

# Weekly evolution digest
/evo-stats --digest --period=7d

# Full evolution report with anomaly detection
/evo-stats --report --include-anomalies --format=markdown

Options & Parameters

Option	Type	Default	Description
`--current`	flag	false	Show only current generation statistics
`--brief`	flag	false	One-line summary of evolution status
`--full-history`	flag	false	Complete evolution history across all generations
`--domains`	flag	false	Quality domain breakdown with per-domain scores
`--trend`	flag	false	Fitness trend analysis with trajectory projection
`--generations`	integer	5	Number of generations to include in trend analysis
`--compare`	flag	false	Compare two generations side-by-side
`--gen1`	integer	none	First generation for comparison
`--gen2`	integer	none	Second generation for comparison
`--format`	string	text	Output format (text, json, table, markdown)
`--output`	string	stdout	Output file path for report generation
`--digest`	flag	false	Generate periodic evolution digest
`--period`	string	7d	Time period for digest generation
`--include-anomalies`	flag	false	Include anomaly detection results
`--report`	flag	false	Generate comprehensive evolution report
`--verbose`	flag	false	Include detailed metrics and diagnostics

Execution Flow

The /evo-stats command follows a structured 5-phase reporting pipeline:

Data Collection: The command queries all evolution data sources including the generation database, quality gate results, fitness evaluation history, and mutation logs. Data is collected from both the runtime system (ETS tables) and persistent storage (Quality DNA files).
Metric Computation: Raw data is transformed into computed metrics including fitness scores per domain, inter-generational deltas, moving averages, trend coefficients, and anomaly scores. All computations are deterministic and reproducible.
Trend Analysis: Time-series analysis is applied to fitness trajectories. The system computes linear regression coefficients for each quality domain, identifies inflection points where fitness gain rates changed, and projects future trajectories based on current momentum.
Anomaly Detection: Statistical anomaly detection identifies outliers in the evolution data. Sudden fitness drops, stalled evolution cycles, or domain-specific regressions are flagged with severity levels (INFO, WARNING, CRITICAL).
Report Synthesis: All computed metrics, trends, and anomalies are synthesized into a structured report. The report format adapts to the requested output format and detail level.

Integration Points

Component	Integration Type	Description
Evolution Engine	Data Source	Generation data, mutation logs, selection results
Quality Gates	Data Source	Gate pass/fail history, domain scores
Quality DNA	Persistence	Cross-session evolution state continuity
SEADF	Framework	Self-evolving framework status and metrics
Telemetry	Monitoring	Evolution metrics event emission
Prismatic Agents	Execution	Evolution-orchestrator agent performs analysis
GitLab API	Reporting	Evolution reports published to GitLab milestones
AIAD Registry	Discovery	Command specification and agent binding

Best Practices

Monitor fitness trends, not absolute scores: A platform at 0.999 fitness has less room for improvement than one at 0.900. Focus on whether the trend is positive and whether gains are distributed across all quality domains rather than concentrated in easy wins.

Use domain breakdown for targeted improvement: When overall fitness stalls, the --domains view reveals which specific quality domains are lagging. This enables targeted evolution campaigns focused on the weakest domains.

Compare generations after major changes: After significant architectural changes or new feature integrations, use --compare to verify that the change did not degrade fitness in unexpected domains. Cross-cutting changes often have non-obvious quality impacts.

Enable anomaly detection in CI/CD: Include /evo-stats --include-anomalies --format=json in CI/CD pipelines to automatically detect evolution regressions. Anomalies with CRITICAL severity should block deployments.

Archive weekly digests: Generate and archive weekly evolution digests (--digest --period=7d) to maintain a historical record of the platform's evolutionary trajectory. This data is invaluable for long-term trend analysis and strategic planning.

Error Handling

Error	Cause	Resolution
`GENERATION_DATA_MISSING`	Requested generation has no stored data	Verify generation number exists (current: Gen 18)
`QUALITY_DNA_UNAVAILABLE`	Quality DNA persistence layer not accessible	Check `.claude/quality-dna/current-state.json`
`METRIC_COMPUTATION_TIMEOUT`	Full history analysis exceeds timeout	Use `--generations` to limit analysis scope
`ANOMALY_DETECTOR_CALIBRATION`	Insufficient data for statistical anomaly detection	Requires minimum 3 generations of data
`FORMAT_UNSUPPORTED`	Requested output format not available	Use supported formats: text, json, table, markdown

Advanced Usage

Custom Metric Dashboards

# Create a focused dashboard for specific domains
/evo-stats --domains=dialyzer,credo,compilation --trend --format=table

# Export evolution data for external visualization
/evo-stats --export --format=csv --include-raw-metrics

# Generate evolution heat map across all domains and generations
/evo-stats --heatmap --generations=all --format=html

Evolution Forecasting

# Project fitness trajectory for next 3 generations
/evo-stats --forecast --horizon=3 --confidence=0.80

# Identify evolution bottlenecks
/evo-stats --bottleneck-analysis --verbose

Integration with Autoevolve

The /evo-stats command is a critical input to the autoevolve system. Evolution statistics inform the mix autoevolve.mega command's strategy selection, helping it decide which quality domains to target and what mutation strategies to apply. When /evo-stats reports fitness stagnation in a domain, the autoevolve system increases mutation pressure in that domain.

Doctrine Compliance

All commands operate under the NO MERCY, NO DOUBTS doctrine:

NO MERCY: Zero tolerance for unmonitored evolution. Every evolution cycle must produce measurable, reported statistics. No blind evolution -- every generation must demonstrate its fitness contribution through quantified metrics.
NO DOUBTS: Full investigation before declaring fitness improvements. Statistics are computed from comprehensive data sources, not sampled or approximated. All reported metrics are deterministic and reproducible.

The evolution statistics system embodies the NABLA Infinity principle of provenance mandatory -- every fitness claim is traceable to specific quality domain measurements, and every trend projection includes explicit confidence intervals and methodology attribution.

/evolve - Living AIAD ecosystem evolution with 5-phase cycle and GitLab observability
/evolve-enforced - Evolution with mandatory QDP reduction
/evolve-patterns - Pattern evolution through meta-evolution analysis
/evolve-quality-gates - Quality gate evolution for warnings, tests and static analysis
/seadf - Self-Evolving Autonomous Development Framework control and monitoring
/quality-gates - Enforce quality gate checkpoints with zero-warning compilation validation
/regression-check - Execute 25 custom Credo regression checks preventing 700+ violations

Connect & Contribute

Created by Tomáš Korcak (korczis) | Open Source under GHL

GitHub | GitLab | LinkedIn | Contact
Developer Portal | Architecture | Meet the Creator

/evo-stats