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.
npx agentmods add agents/mickeyyaya/evolve-loop/agent-templatesgit clone --depth 1 https://github.com/mickeyyaya/evolve-loopWhat 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 | $0.00013 | $0.04206 |
| Opus 5 | $0.00006 | $0.02103 |
| Sonnet 5 | $0.00003 | $0.00841 |
| Haiku 4.5 | $0.00001 | $0.00421 |
Grade A, and why
agent-templates 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 2d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 328 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Templates — Shared Schemas
v12.0.0 status:
legacy/scripts/...paths referenced below were removed in the v12 flag day. Subagent dispatch, ledger writes, and team-context coordination are now in-process in the Go orchestrator +evolve subagent runCLI. Treat bash snippets as contracts; do not invoke them directly.
Shared input/output schemas for evolve-loop agents. Each agent references this file instead of duplicating boilerplate. Agent-specific fields are documented in the individual agent files.
Agent Definition Schema
Every agent file MUST include these frontmatter fields (in addition to name, description, model, capabilities, tools):
| Field | Type | Required | Description |
|---|---|---|---|
perspective |
string | yes | The evaluative lens through which this agent interprets every input. Sets the default bias for all judgments. Example: "adversarial reviewer seeking failure modes". |
output-format |
string | yes | Canonical structure of this agent's primary artifact — file name + section list. Enables deterministic merging in parallel fan-out scenarios. Example: "audit-report.md — Verdict, Defect Table, Eval Gate result". |
Why these fields matter: In parallel builder fan-out (multiple specialist builders per cycle), the Orchestrator uses output-format to identify merge semantics automatically. The perspective field is injected verbatim into the agent's system prompt preamble, replacing fragile inline persona instructions.
Shared Context Block
All agents receive a JSON context block with these common fields:
| Field | Type | Description |
|---|---|---|
cycle |
number | Current cycle number |
workspacePath |
string | Path to .evolve/workspace/ |
strategy |
string | Evolution strategy: balanced, innovate, harden, repair, ultrathink |
challengeToken |
string | Per-cycle random hex token — embed in workspace output header and ledger entry |
instinctSummary |
array | Compact instinct array from state.json (inline) |
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.
- 2d ago First seen · 328 lines · 13 tokens per session scan A 35c8a1c140a8
agent-templates is an agent published in the GitHub repository mickeyyaya/evolve-loop (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 13 tokens to every session and 4,206 once invoked, about $0.0001 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 2026-08-31.
Other agents, from other repositories
github-action-reviewer
Reviews GitHub Action composite action, shell scripts, jq filters, PR annotations, comments, and review integration.
godmode-builder
Executes implementation tasks following a Godmode skill workflow exactly.
lead
Workflow orchestrator. Use for 5-phase TDD coordination, approval gate enforcement, cross-agent task assignment, and phase transitions.
replanner
Triggered by failure-classifier on F2-F4 escalations. Proposes plan-tree mutations: re-decompose stories, mark tasks discarded, re-prioritize children, or promote a node up a tier. Read-only on code; mutations applied via master-planner.
Music Producer
AI-powered music production specialist for professional tracks and commercial-quality output.
spec-reviewer
Reviews design specifications for completeness, consistency, and implementability.