self-improve

self-improve is a skill for Claude Code, Codex from sskarz/nanoclawbster. It costs 50 tokens per session (1,791 once invoked), scanned A, original, MIT.

A development workflow for changing the NanoClawbster codebase in an isolated workspace, then sending the changes through pull-request checks and deployment approval.

In plain words
What is it for?
Writing features, fixing bugs, creating branches and pull requests, running GitHub Actions checks, and deploying after the user approves.
Why use it?
It separates live code from editable work and provides a defined path from a feature or bug fix to review and deployment.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: names the TodoWrite tool.

Good fit Writing features, fixing bugs, creating branches and pull requests, running GitHub Actions checks, and deploying after the user approves.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/sskarz/nanoclawbster/self-improve
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 sskarz/nanoclawbster --skill self-improve
Clone the repo
git clone --depth 1 https://github.com/sskarz/nanoclawbster

Made for: Claude Code, 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 self-improve

README.md
[![agentmods](https://agentmods.dev/badge/skills/sskarz/nanoclawbster/self-improve/github.svg)](https://agentmods.dev/skills/sskarz/nanoclawbster/self-improve)
Your own site
<a href="https://agentmods.dev/skills/sskarz/nanoclawbster/self-improve"><img src="https://agentmods.dev/badge/skills/sskarz/nanoclawbster/self-improve/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 self-improve

Your own site · 80×15
<a href="https://agentmods.dev/skills/sskarz/nanoclawbster/self-improve"><img src="https://agentmods.dev/badge/skills/sskarz/nanoclawbster/self-improve.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 50 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,791 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 $0.00050 $0.01791
Opus 5 $0.00025 $0.00896
Sonnet 5 $0.00010 $0.00358
Haiku 4.5 $0.00005 $0.00179

Measured 10d ago against content hash 48034889c87a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

self-improve 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 10d 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.

container/skills/self-improve/SKILL.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Self-Improvement Workflow

Write features, fix bugs, and create PRs for the NanoClawbster codebase. You work in an isolated dev workspace, push via Composio (GitHub Actions runs CI), and deploy after user approval.

Workspace Layout

/workspace/project        → Live code (READ-ONLY)
/workspace/dev/nanobot    → NanoClawbster git clone (edit here for nanobot changes)
/workspace/dev/<repo>/    → Other repos (clone with: git clone <url> /workspace/dev/<name>)
/workspace/group          → Group files (READ-WRITE)

Never edit /workspace/project directly. All changes happen in /workspace/dev/nanobot.

Step-by-Step Workflow

0. Announce and Track

Before doing anything else:

  1. Use send_message to tell the user what you're working on:

    "Working on: [brief description of the task]. I'll send you the PR link when done."

  2. Use TodoWrite to set up your full task list with ALL steps below as pending todos. This keeps you on track and lets the user see progress. Include a todo for creating the PR — this is mandatory.

Example todos:

  • Sync dev workspace and create branch
  • Read and understand relevant files
  • Make the code changes
  • Run local build check (tsc --noEmit)
  • Push via Composio and create PR ← never skip this
  • Verify CI passes on the PR

Mark each todo in_progress before starting it, completed immediately when done.

1. Sync Dev Workspace

cd /workspace/dev/nanobot
git checkout main
git pull origin main
npm install

If git pull fails because origin points to the local host path, fix it first:

git remote set-url origin https://github.com/sskarz/nanoclawbster.git
git pull origin main

2. Create a Branch

cd /workspace/dev/nanobot
git checkout -b feature/your-change-name

This branch is local only — Composio handles the GitHub branch when you push.

3. Read Before Editing

Before making changes, read the files you'll edit. Understand the existing patterns. Do not modify code you haven't read.

Read the full file on GitHub · 177 lines

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. 10d ago First seen · 177 lines · 50 tokens per session scan A 48034889c87a

Subscribe to this mod's changes

self-improve is a skill published in the GitHub repository sskarz/nanoclawbster (2 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 1,791 once invoked, about $0.0003 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.