agentmods.dev

control

control is a skill for Codex from BackToCimaCoppi/Praxis. It costs 58 tokens per session (6,307 once invoked), scanned A, original, Apache-2.0.

A manually triggered controller for advancing one specified task in a project’s active task-control documents.

In plain words
What is it for?
Use it with an explicit task key to read the needed project context, complete only that task, record its status, and stop.
Why use it?
It prevents accidental work on neighboring tasks and keeps each task’s outputs and status within its assigned boundaries.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: mentions AGENTS.md.

not rated 6repo changed 29d ago A scan 58 tokens original Apache-2.0

Good fit Use it with an explicit task key to read the needed project context, complete only that task, record its status, and stop.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/backtocimacoppi/praxis/control
View source ↗ BackToCimaCoppi/Praxis
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.

Any agent
npx skills add BackToCimaCoppi/Praxis --skill control
Clone the repo
git clone --depth 1 https://github.com/BackToCimaCoppi/Praxis

Made for: Codex.

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 control

README.md
[![agentmods](https://agentmods.dev/badge/skills/backtocimacoppi/praxis/control/github.svg)](https://agentmods.dev/skills/backtocimacoppi/praxis/control)
Your own site
<a href="https://agentmods.dev/skills/backtocimacoppi/praxis/control"><img src="https://agentmods.dev/badge/skills/backtocimacoppi/praxis/control/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for control

Your own site · 80×15
<a href="https://agentmods.dev/skills/backtocimacoppi/praxis/control"><img src="https://agentmods.dev/badge/skills/backtocimacoppi/praxis/control.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,307 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00058 $0.06307
Opus 5.5 $0.00023 $0.02523
Sonnet 5.5 $0.00012 $0.01261
Haiku 4.5 $0.00006 $0.00631

Measured 29d ago against content hash 9d2c34038cad, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

control 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 29d ago.

The scan reads SKILL.md. This mod also ships 11 executable files (scripts/_common.py, scripts/archive_control.py, scripts/bootstrap_project.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/control/SKILL.md · 348 lines

How it starts

The opening of the file, as written. The whole thing — 348 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Control

按当前项目 docs/00-任务总控/ 下的总控文档推进任务。本 skill 是通用方法论 skill（用户级），不绑定具体项目。

真值源：references/总控规范.md（本 skill 内）。

🛑 执行边界（最高红线，永远遵守）

这是本 skill 最重要的一条。所有其他规则都让位于此。

  1. 用户指定 $control <key> Tn → 只执行 Tn（支持 Tn.x 二级子任务）。不做 Tn-1，不做 Tn+1，不做同级兄弟，不"顺手"做相关任务。
  2. Tn 的「输出物」全部产出 → 立刻停止，回填状态，向用户报告。等用户下一步指令。
  3. ✅ 读不设禁区：可以读其他子任务详情与产出物、总控任何部分、仓库任何文件——开工前把背景挖够再动手（读什么放开 ≠ 做什么放开）
  4. ❌ 不可写入其他子任务范围（即使发现"很容易顺便做"也不要做）
  5. 如果发现 Tn 的范围定义有问题（应扩大或缩小） → 停止施工，向用户报告问题，不要自行扩张。但**"工作包某句描述过时了"不是范围问题**——与已冻结上游冲突时，按最新冻结真值修订描述、留痕、继续跑（总控规范 §4.6）
  6. 自主补读亲自读：为补背景直接读文件 / 检索即可，禁止为"补背景"派子 agent / 起深度调查（读是加法、派 agent 是乘法）；背景缺口大到需要专门调查 → 按第 5 条停机上报，不自己扩编
  7. 过程资产边界：执行 Tn 时产生的中间文档（设计稿、清单、批判稿等），只能写到 _shared/ 且文件名带 T{n}- 前缀（任务级共享资产无前缀）；不可用其他子任务的 T{m}- 前缀（m≠n），不可平铺到任务根目录，不可新建 _T{n}/ 目录。详见总控规范 §1.1.1。
  8. 规则铸造权不在工作包手上：只有项目级指令文档与 skill 能铸造规则。工作包 / 评审报告 / goal 章程里的「硬门」「死亡线」「必须审批」一律是执行建议，不具停机权；死亡线是封闭清单，任何人无权新增——包括你（总控规范 §4.4）
  9. 停机白名单封闭，只有三条：①两份都已冻结的真值真矛盾且改变业务结果 ②命中项目死亡线清单的业务规则要变 ③不可逆 / 外发动作。其余一律先修后报——笔误、脚本 bug、过期描述、证据充分的简单缺陷，在预授权边界内修复 + 留痕 + 继续。减少临时协调不覆盖项目声明的授权角色、环境占用、职责分离或强制检查点。冲突批处理与 EXECUTION_BLOCKED 细则以总控规范 §4.5 为真值；goal 执行态以章程 §4 白名单为准（章程已由对应权限角色预授权的动作不触发第 ③ 条）
  10. 获批总控可包含本地提交授权：只有经任务治理负责人批准的总控，才预授权在其范围内 git commit；项目补丁可要求额外检查。推送远端仍需发布授权人显式授权（总控规范 §4.7）
  11. 写入范围必须机器核对：开工记录 start_commit；工作包列出 allowed_write_paths 与精确到 Markdown 标题的 allowed_cross_task_writes；交付前只对本子任务明确列出的候选提交运行 check_write_scope.py。无关提交不阻断，候选提交夹带越界文件直接失败（总控规范 §4.8）
  12. 长程 goal 必须磁盘复水：首次启动、--resume、新会话或自动上下文压缩后，第一次写入前重读当前工作包、全部强制阅读、章程、goal 断点与飞行日志尾部；连续未压缩轮次不机械复读整套 skill（总控规范 §4.9）

项目约定

本 skill 假设当前项目遵循以下固定路径约定（所有项目通用，不带任何具体项目烙印）：

路径 用途
<PROJECT_ROOT>/docs/00-任务总控/ 总控目录树根
<PROJECT_ROOT>/docs/00-任务总控/README.md 活跃任务索引
<PROJECT_ROOT>/docs/00-任务总控/归档/ 已完成任务归档树
<PROJECT_ROOT>/docs/00-任务总控/归档/README.md 归档总索引
<PROJECT_ROOT>/docs/00-任务总控/归档/V{x}/README.md 版本归档索引
<PROJECT_ROOT>/docs/00-任务总控/{YYYY-MM-DD}-{任务名}/ 任务子目录（日期前缀防多 worktree 编号撞车）
<PROJECT_ROOT>/docs/00-任务总控/{YYYY-MM-DD}-{任务名}/README.md 任务主总控（拆分 / 单文件统一）
<PROJECT_ROOT>/docs/00-任务总控/{YYYY-MM-DD}-{任务名}/T{n}-{子任务名}.md 拆分模式的一级子任务工作包
<PROJECT_ROOT>/docs/00-任务总控/{YYYY-MM-DD}-{任务名}/T{n}/T{n}.md 被拆过的父任务说明（保留背景/范围，原 T{n}-*.md 迁移到此）
<PROJECT_ROOT>/docs/00-任务总控/{YYYY-MM-DD}-{任务名}/T{n}/T{n}.{m}-{子任务名}.md 二级子任务工作包（split 后产生）
<PROJECT_ROOT>/.agents/local/active-control per-worktree 激活配置（gitignored）
<PROJECT_ROOT>/.gitignore 必须包含 .agents/local/

Read the full file on GitHub · 348 lines

Files

What ships with it

13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 29d ago Changed 9d2c34038cad
  2. 1mo ago First seen · 348 lines · 58 tokens per session scan A 4301debbbed6

Subscribe to this mod's changes

control is a skill published in the GitHub repository BackToCimaCoppi/Praxis (6 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 58 tokens to every session and 6,307 once invoked, about $0.0002 per session on Opus 5.5. 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.

Claude CodeLLMPlanningai-developmentclaude-skillsdeveloper-toolsmethodology
Related

Other skills, from other repositories

prompt-master

Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…

nidhinjs/prompt-master · 78 tokens

browser-tools

Safety rules over the upstream agent-browser skill (scraping ethics, auth, request pacing, local URLs). URL policy is the Claude Code sandbox's job when the operator enables it. Use for browser automation.

yonatangross/orchestkit · 44 tokens

rag-retrieval

Retrieval-Augmented Generation patterns for grounded LLM responses. Use when building RAG pipelines, embedding documents, implementing hybrid search, contextual retrieval, HyDE, agentic RAG, multimodal RAG, query decomposition, reranking, or pgvector search.

yonatangross/orchestkit · 58 tokens

architecture-patterns

Architecture validation and patterns for clean architecture, backend structure enforcement, project structure validation, test standards, and context-aware sizing. Use when designing system boundaries, enforcing layered architecture, validating project structure, defining test standards, or choosing the right…

yonatangross/orchestkit · 56 tokens

auto

Intent-classified router, the front door to OrchestKit and the DEFAULT entry point for any goal-shaped request. Classifies a plain-English goal and routes it to the right specialist skill. Routing is never overhead, so use it even when the target skill seems obvious; skip only when already executing inside another…

yonatangross/orchestkit · 99 tokens

langgraph

LangGraph 1.x (LTS) Python workflow patterns for state management, delta channels, resilience (node timeouts, error handlers, graceful drain), routing, parallel execution, supervisor-worker, tool calling, checkpointing, human-in-loop, streaming (v2 format), subgraphs, and functional API. Use when building LangGraph…

yonatangross/orchestkit · 80 tokens