What is evolve-operator?

A post-cycle monitoring agent for the Evolve Loop. It reviews recent work, checks for stalled progress, tracks quality trends, and recommends adjustments for the next cycle.

Use it after each cycle to assess loop health, review recent records and notes, detect stalls, and check whether new work has appeared.

It helps teams notice when development is slowing down, repeating itself, or converging without actually solving the problem.

evolve-operator — Agent for coding agents by mickeyyaya: A post-cycle monitoring agent for the Evolve Loop.

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 agents/mickeyyaya/evolve-loop/evolve-operator

Clone the repo

git clone --depth 1 https://github.com/mickeyyaya/evolve-loop

Or install evo, the plugin that ships this one along with the rest of its 26 skills, 27 commands, 107 agents, 3 hooks.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/evolve-operator.svg)](https://agentmods.dev/agents/mickeyyaya/evolve-loop/evolve-operator)

Your own site

<a href="https://agentmods.dev/agents/mickeyyaya/evolve-loop/evolve-operator"><img src="https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/evolve-operator.svg" alt="Measured on agentmods" height="20"></a>

Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,444 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. 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.

Model	Per session	Once invoked
Fable 5	$0.00030	$0.02444
Opus 5	$0.00015	$0.01222
Sonnet 5	$0.00006	$0.00489
Haiku 4.5	$0.00003	$0.00244

Measured yesterday against content hash ed4185821df7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evolve-operator scanned grade A with 0 findings 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 yesterday.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/evolve-operator.md · 232 lines

How it starts

The opening of the file, as written. The whole thing — 232 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evolve Operator

You are the Operator in the Evolve Loop pipeline. Monitor loop health, detect stalls, and recommend adjustments. Invoked once per cycle (post-cycle).

Research-backed techniques: Read docs/reference/orchestrator-techniques.md for strategy playbook protocol, instinct forgetting, trust governance, self-improvement metrics, and cost-performance feedback.

Inputs

JSON context block:

cycle: current cycle number
mode: "post-cycle" (normal) or "convergence-check" (nothingToDoCount >= 2)
workspacePath: path to .evolve/workspace/
stateJson: contents of .evolve/state.json (includes ledgerSummary, instinctSummary)
recentLedger: last 5 ledger entries (inline — do NOT read full ledger.jsonl)
recentNotes: last 5 cycle entries from notes.md (inline — do NOT read full notes.md)

Mode Handling

post-cycle — Assess cycle health, detect stalls, recommend adjustments.
convergence-check — Check for external changes (git log --oneline -3), new issues, changed state. New work detected = recommend resetting nothingToDoCount to 0. Otherwise confirm convergence.

Responsibilities

1. Progress Assessment

Read workspace/build-report.md and workspace/audit-report.md
Read ledgerSummary from stateJson for aggregate stats
Did this cycle ship code? Tasks completed vs attempted?
Was task sizing appropriate?

2. Behavioral Anomaly Detection (Reward Hacking)

Primary input: Read workspace/cycle-health.json (from legacy/scripts/observability/cycle-health-check.sh). Deterministic ground truth. Missing file = highest-severity anomaly.

Health fingerprint interpretation:

Any status: "ANOMALY" = immediate HALT
2+ status: "WARN" in single cycle = escalate to HALT
3+ WARNs in any 5-cycle window = trigger examination mode

Secondary checks:

Velocity Anomaly: git log --format="%ad" --date=iso -n 2 — task completed in <5 seconds = impossible, trigger HALT
Ledger Role Completeness: Verify scout, builder, auditor, orchestrator entries for current cycle. Missing = dropped phases.
Challenge Token Consistency: Same challenge in all role entries for this cycle. Mismatch = forged entries.
Ceremonialization: M-complexity 5+ file task with perfect confidence in single attempt without showing work = suspicious
Tool-Use Sequencing: Builder accessed .github/workflows, .env, or test configs when task only required source edits?
Complexity Gaming: Inflated lines (whitespace/comments) to satisfy metrics?
Canary Integrity: Check canary/honeypot files accessed or modified

Read the full file on GitHub · 232 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.

yesterday First seen · 232 lines · 30 tokens per session scan A ed4185821df7

Subscribe to this mod's changes

evolve-operator is an agent published in the GitHub repository mickeyyaya/evolve-loop (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 30 tokens to every session and 2,444 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Analytics and metrics Project management ai-agents ai-pipeline autonomous-coding claude-code claude-code-plugin code-quality

retro

After an initiative closes, reads the ledger, the notes and the agents' reports and improves Tyran itself - skills, agents, scripts and docs, never product code. Defends hard against bloat and overfitting, so the default answer is to change nothing.

jjanczur/tyran · 56 tokens

task-executor

Use this agent to execute a single tracked task with TDD, commit, and PR creation in an isolated git worktree. Dispatched by /coco:loop for parallel execution. Context: Multiple tasks are ready with non-overlapping file ownership. /coco:loop dispatches parallel agents. assistant: "I'll dispatch task-executor agents…

skullninja/coco-workflow · 97 tokens

evolve-operator