evolve-agents

evolve-agents is a command for Claude Code from TheLobbi/claude. It costs 0 tokens per session (3,088 once invoked), scanned B, original, MIT.

A command for reviewing how agents perform and identifying missing abilities or possible improvements. It can examine trends, capability gaps, suggestions, prompt history, and prompt experiments.

In plain words
What is it for?
Use it to analyze one agent or all agents, create weekly reports, check improvement thresholds, review prompt history, and compare prompt changes.
Why use it?
It gives a structured way to find why an agent is underperforming and decide what to improve.

Command for Claude Code

Part of the claude-orchestration plugin — 49 commands, 9 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/thelobbi/claude/evolve-agents
Clone the repo
git clone --depth 1 https://github.com/TheLobbi/claude

Made for: Claude Code.

Or install claude-orchestration, the plugin that ships this one along with the rest of its 49 commands, 9 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for evolve-agents

README.md
[![agentmods](https://agentmods.dev/badge/commands/thelobbi/claude/evolve-agents.svg)](https://agentmods.dev/commands/thelobbi/claude/evolve-agents)
Your own site
<a href="https://agentmods.dev/commands/thelobbi/claude/evolve-agents"><img src="https://agentmods.dev/badge/commands/thelobbi/claude/evolve-agents.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,088 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.03088
Opus 5 $0.00000 $0.01544
Sonnet 5 $0.00000 $0.00618
Haiku 4.5 $0.00000 $0.00309

Measured 3d ago against content hash ab2759c66241, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

evolve-agents scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Asks the agent to reveal its instructionsmediumSystem prompt leakage

Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.

Display prompt version history for an agent:
.claude/commands/evolve-agents.md · 495 lines

How it starts

The opening of the file, as written. The whole thing — 495 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evolve Agents - Agent Self-Improvement Analysis

Analyze agent performance, identify capability gaps, and propose improvements using the Evolution System.

Usage

# Analyze all agents
.claude/commands/evolve-agents.md analyze

# Analyze specific agent
.claude/commands/evolve-agents.md analyze --agent-id=coder-agent

# Generate weekly evolution report
.claude/commands/evolve-agents.md report --period=weekly

# Check thresholds and recommend updates
.claude/commands/evolve-agents.md check-thresholds

# View capability gaps
.claude/commands/evolve-agents.md gaps

# View skill suggestions
.claude/commands/evolve-agents.md suggestions

# View prompt history
.claude/commands/evolve-agents.md prompt-history --agent-id=coder-agent

# Evolve agent prompts
.claude/commands/evolve-agents.md evolve --agent-id=coder-agent

# Run A/B test on prompts
.claude/commands/evolve-agents.md ab-test --agent-id=coder-agent

Command Actions

1. analyze - Comprehensive Agent Analysis

Analyze agent performance metrics including:

  • Success rate over time
  • Average task duration
  • Token efficiency (success per 1000 tokens)
  • User feedback ratings
  • Performance trends (improving, stable, declining)

Output:

  • Performance summary table
  • Trend analysis
  • Comparison with previous periods
  • Recommendations for improvement

Example Output:

Agent Performance Analysis
========================

Agent: coder-agent
------------------
Success Rate:        87.5% (↑ 5.2% vs last week)
Avg Duration:        45.3s (↓ 8.1s vs last week)
Token Efficiency:    0.72 (↑ 0.08 vs last week)
User Rating:         4.2/5.0
Performance Trend:   IMPROVING

Tasks Completed:     120
Failures:           15
Retries:            8

Top Error Types:
1. timeout (5 occurrences)
2. validation_error (4 occurrences)
3. tool_limitation (3 occurrences)

2. report - Generate Evolution Report

Generate a comprehensive evolution report for the specified period:

  • Overall system performance
  • Per-agent performance summaries
  • Identified capability gaps
  • Skill suggestions
  • Prompt updates
  • Improvements deployed

Read the full file on GitHub · 495 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 495 lines · 0 tokens per session scan B ab2759c66241

Subscribe to this mod's changes

evolve-agents is a command published in the GitHub repository TheLobbi/claude (21 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,088 tokens. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.