How do I install the Project Upgrader skill?

Run npx skillmds@latest add rumotion/project-upgrader in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Project Upgrader safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Project Upgrader?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Project Upgrader free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Project Upgrader?

rumotion (@rumotion) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Project Upgrader by rumotion — Agent Skill

Name: Project Upgrader
Author: rumotion

Project Upgrader

Upgrades an existing target project to the latest AI Agent Project Template version. Use when asked to upgrade another project, port template updates, or bring an existing repository up to date with the latest template logic, skills, workflows, adapters, and validation tools without altering or breaking existing project scripts, tests, assets, or domain logic.

rumotion 3779a40 3.5 KB Updated Sep 17, 2026

File contents

Project Upgrader Skill

Use this skill whenever asked to upgrade an existing repository/project using the latest version of the AI Agent Project Template.

Core Directives

[!IMPORTANT]

Zero Damage to Existing Project Domain Code: Never overwrite, delete, or alter custom application code, test scripts (test_*.py, *.spec.ts, etc.), project assets, extraction logic, authentication configs (*.json, .env), or custom domain skills (e.g., kumo-canvas).

Preserve Custom AGENTS.md Rules: Always retain project-specific domain rules found in the target repository's AGENTS.md (e.g. database safety rules, test board rules).

Context Budget Enforcement: Ensure target AGENTS.md stays <= 6,500 chars and memory-bank/handoff.md stays <= 1,200 chars to satisfy FAST_INIT budget checks.

Upgrade Execution Procedure

Step 1: Preflight & Dry-Run

Identify target project path (e.g. c:\AI\_code\InnovationLab\KUMO\KumoQAToolkit).

Run a dry-run to preview planned modifications:

python scripts/upgrade-target.py --target <target-path> --dry-run

Read target AGENTS.md and .agents/skills/ to verify custom domain rules and custom domain skills to preserve.

Step 2: Automated Framework Upgrade

Execute the automated upgrade tool:
```
python scripts/upgrade-target.py --target <target-path>
```
This automatically:
- Preserves custom domain rules in AGENTS.md while merging latest canonical instructions, falsifiable predictions, escalation ladder, and bimodal execution.
- Synchronizes multi-agent adapters (GEMINI.md, CLAUDE.md, CONVENTIONS.md, .windsurfrules, .codex/AGENTS.md, .cursor/rules/agents.mdc, .github/copilot-instructions.md, .clinerules/, .agents/rules/).
- Synchronizes 10 workflows across workflows/, .agents/workflows/, .clinerules/workflows/.
- Synchronizes standard skills across .agents/skills/, .claude/skills/, .cline/skills/ while preserving custom domain skills intact.
- Copies safety hooks, reviewer contracts, documentation, and validation scripts (scripts/check-template.py, scripts/init-fast.py, scripts/detach-remote.py, scripts/benchmark-context.py, scripts/upgrade-target.py).
- Updates Memory Bank schemas to schema v2 while keeping project history.

Step 3: Verification & Health Check

Run fast template validation in target:

python <target-path>/scripts/check-template.py --fast

Run benchmark self-test and hook verification in target:

python <target-path>/scripts/benchmark-context.py --self-test
python <target-path>/scripts/hooks/verify-fixtures.py

Verify target application test files compile/pass cleanly:
```
python -m py_compile <test_files>
```

Step 4: Commit Upgraded Target

Stage and commit upgraded framework files in target repository:

git add AGENTS.md GEMINI.md CLAUDE.md CONVENTIONS.md .codex/ .cursor/ .github/ .clinerules/ .agents/ .claude/ .gemini/ docs/ workflows/ scripts/ benchmarks/ memory-bank/ .gitignore
git commit -m "feat(template): upgrade framework to latest AI Agent Project Template"

---name: project-upgraderdescription: Upgrades an existing target project to the latest AI Agent Project Template version. Use when asked to upgrade another project, port template updates, or bring an existing repository up to date with the latest template logic, skills, workflows, adapters, and validation tools without altering or breaking existing project scripts, tests, assets, or domain logic.---# Project Upgrader SkillUse this skill whenever asked to upgrade an existing repository/project using the latest version of the AI Agent Project Template.## Core Directives> [!IMPORTANT]> 1. **Zero Damage to Existing Project Domain Code**: Never overwrite, delete, or alter custom application code, test scripts (`test_*.py`, `*.spec.ts`, etc.), project assets, extraction logic, authentication configs (`*.json`, `.env`), or custom domain skills (e.g., `kumo-canvas`).> 2. **Preserve Custom AGENTS.md Rules**: Always retain project-specific domain rules found in the target repository's `AGENTS.md` (e.g. database safety rules, test board rules).> 3. **Context Budget Enforcement**: Ensure target `AGENTS.md` stays <= 6,500 chars and `memory-bank/handoff.md` stays <= 1,200 chars to satisfy `FAST_INIT` budget checks.---## Upgrade Execution Procedure### Step 1: Preflight & Dry-Run1. Identify target project path (e.g. `c:\AI\_code\InnovationLab\KUMO\KumoQAToolkit`).2. Run a dry-run to preview planned modifications: ```bash python scripts/upgrade-target.py --target <target-path> --dry-run ```3. Read target `AGENTS.md` and `.agents/skills/` to verify custom domain rules and custom domain skills to preserve.### Step 2: Automated Framework Upgrade1. Execute the automated upgrade tool: ```bash python scripts/upgrade-target.py --target <target-path> ``` This automatically: - Preserves custom domain rules in `AGENTS.md` while merging latest canonical instructions, falsifiable predictions, escalation ladder, and bimodal execution. - Synchronizes multi-agent adapters (`GEMINI.md`, `CLAUDE.md`, `CONVENTIONS.md`, `.windsurfrules`, `.codex/AGENTS.md`, `.cursor/rules/agents.mdc`, `.github/copilot-instructions.md`, `.clinerules/`, `.agents/rules/`). - Synchronizes 10 workflows across `workflows/`, `.agents/workflows/`, `.clinerules/workflows/`. - Synchronizes standard skills across `.agents/skills/`, `.claude/skills/`, `.cline/skills/` while preserving custom domain skills intact. - Copies safety hooks, reviewer contracts, documentation, and validation scripts (`scripts/check-template.py`, `scripts/init-fast.py`, `scripts/detach-remote.py`, `scripts/benchmark-context.py`, `scripts/upgrade-target.py`). - Updates Memory Bank schemas to schema v2 while keeping project history.### Step 3: Verification & Health Check1. Run fast template validation in target: ```bash python <target-path>/scripts/check-template.py --fast ```2. Run benchmark self-test and hook verification in target: ```bash python <target-path>/scripts/benchmark-context.py --self-test python <target-path>/scripts/hooks/verify-fixtures.py ```3. Verify target application test files compile/pass cleanly: ```bash python -m py_compile <test_files> ```### Step 4: Commit Upgraded Target1. Stage and commit upgraded framework files in target repository: ```bash git add AGENTS.md GEMINI.md CLAUDE.md CONVENTIONS.md .codex/ .cursor/ .github/ .clinerules/ .agents/ .claude/ .gemini/ docs/ workflows/ scripts/ benchmarks/ memory-bank/ .gitignore git commit -m "feat(template): upgrade framework to latest AI Agent Project Template" ```

Frequently asked questions