iquestlab/modularrsi/kernel-boundaries

Claude Code and agent skill by iquestlab/modularrsi

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Which files are Kernel (NEVER edit, off-limits) versus which are evolvable modules. Read this BEFORE making any edits.

Install

npx vskill@latest install iquestlab/modularrsi/kernel-boundaries

SKILL.md

Kernel boundaries

The codebase you're editing is split into Kernel (fixed plumbing) and Modules (evolvable strategy). You MAY only modify modules.

Kernel — DO NOT EDIT

Anything at the top level of staging/ that is NOT a modules/<type>/ file:

File / DirWhat it doesWhy off-limits
protocols.pyDefines the Protocol contracts every module must satisfyChanging it would invalidate every existing module's interface
services.pyKernelServices: logger / failure reporter / stats writer / ATIF trajectory recorderWired up at startup; modules consume but don't define these
agent.pyTerminus2Modular / Terminus2ModularEditor classes; agent lifecycleThis file is what loads YOU
library.pyAuto-discovers modules (so you do NOT need to update imports here when adding a new module)Auto-discovery means you don't need to touch it
composer/The Composer that picks which module variant to use PER TASK (reads the task + each variant's DESCRIPTION)llm_dynamic per-task selection; off-limits to you — you grow the library, it selects
__init__.pyPackage exports—

The editor tooling will REJECT any <edit_file> or <create_file> action that targets these files with path is off-limits (Kernel).

Modules — SAFE to edit

Everything under modules/. Five module types, each with sub-directories:

modules/
├── agent_loop/ ← main control loop incl. planning strategy (ReAct / plan-execute / ToT / ...)
├── observation/ ← env raw output → agent-readable obs
├── context_mgmt/ ← summarize / unwind / compress chat history
├── tools/ ← what actions the agent can take (tmux send-keys etc.)
└── verification/ ← decide when agent has finished the task

You can:

  • Modify a routed, non-baseline variant when the action explicitly permits

it. Every baseline.py is read-only; replacing baseline behavior requires a new variant.

  • Create new .py files under any modules/<type>/ directory.
  • New files are auto-discovered by library.py on next startup — you do NOT

need to register them anywhere.

You CANNOT:

  • Modify Kernel files (above).
  • Delete files (tooling has no delete action).
  • Rename files (no rename action).
  • Add Python dependencies (no way to touch pyproject.toml).

Quick reference

If you ever wonder "is this file editable?":

  • Path starts with modules/<type>/, ends in .py, and is not a baseline or

other protected file? → YES, subject to the routed action.

  • Otherwise → NO, off-limits.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/iquestlab/modularrsi/kernel-boundaries/badge)](https://verified-skill.com/skills/iquestlab/modularrsi/kernel-boundaries)

Something wrong with this skill or its page? Tell us.