---
title: "high-agency"
description: "Lightweight high-agency coding with impact calibration, adaptive model routing, targeted verification, and bounded autonomous iteration for Codex."
canonical: https://agentpluginsdirectory.com/plugins/high-agency
last-updated: 2026-09-25
---

# high-agency
Lightweight high-agency coding with impact calibration, adaptive model routing, targeted verification, and bounded autonomous iteration for Codex.
- Slug: high-agency
- Publisher: Ssojux2
- Repository: https://github.com/Ssojux2/high-agency-codex
- Manifest: plugins/high-agency/plugin.json
- Version: 0.9.0
- License: MIT
- Category (editorial): other
- Skills: 3 (bounded-autonomy, high-agency-coding, high-agency-doctor)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-09-21
- Repository last pushed: 2026-09-23
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/high-agency
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What high-agency does, in the publisher's words

Lightweight Codex plugin containing:

- high-agency-coding: direct autonomous coding with fresh verification.
- bounded-autonomy: opt-in bounded continuation via the bundled Stop hook.

See the repository root README for installation and usage.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/Ssojux2/high-agency-codex/HEAD/plugins/high-agency/README.md

## Skills

- bounded-autonomy: Use when the user asks Codex to keep working autonomously, iterate until done, fix until tests pass, or run a bounded Ralph-style loop with objective or practically verifiable acceptance criteria.
- high-agency-coding: Use when implementing, debugging, refactoring, reviewing, or planning code where autonomous execution and evidence-based verification are useful; defaults to single-agent execution and escalates models, reasoning, verification, or review only when they materially improve correctness or efficiency.
- high-agency-doctor: Use when diagnosing High Agency installation, routing, hooks, model/effort configuration, or unexpected delegation behavior in Codex. Performs read-only checks and never modifies project files.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.