kube-coder: Skill for Claude Code

.claude/skills/kc-entrypoint-change/SKILL.md

kc-entrypoint-change is a skill for Claude Code from imran31415/kube-coder. It costs 75 tokens per session (1,722 once invoked), scanned C, original, MIT.

A safety guide for editing the startup scripts that boot kube-coder workspace pods. A pod is a deployed application environment that runs the workspace.

In plain words
What is it for?
Use it when changing startup credentials, services, terminal connections, SSH settings, or other pod-boot behavior.
Why use it?
Startup mistakes affect every workspace after restart, and the setup uses separate persistent and temporary home directories plus Helm templates.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md; mentions OpenCode.

This is imran31415/kube-coder's own configuration. It tells Claude Code how to work on kube-coder itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kube-coder configures →

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /home/ubuntu/.ssh.

Reuse

Borrowing it

Nothing to install: this file belongs to imran31415/kube-coder. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/imran31415/kube-coder/main/.claude/skills/kc-entrypoint-change/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/imran31415/kube-coder

Made for: Claude Code.

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 kc-entrypoint-change

README.md
[![agentmods](https://agentmods.dev/badge/skills/imran31415/kube-coder/kc-entrypoint-change/github.svg)](https://agentmods.dev/skills/imran31415/kube-coder/kc-entrypoint-change)
Your own site
<a href="https://agentmods.dev/skills/imran31415/kube-coder/kc-entrypoint-change"><img src="https://agentmods.dev/badge/skills/imran31415/kube-coder/kc-entrypoint-change/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 kc-entrypoint-change

Your own site · 80×15
<a href="https://agentmods.dev/skills/imran31415/kube-coder/kc-entrypoint-change"><img src="https://agentmods.dev/badge/skills/imran31415/kube-coder/kc-entrypoint-change.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,722 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. 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 $0.00075 $0.01722
Opus 5.5 $0.00030 $0.00689
Sonnet 5.5 $0.00015 $0.00344
Haiku 4.5 $0.00007 $0.00172

Measured 13d ago against content hash 1ae8bcfaf963, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

kc-entrypoint-change scanned grade C with 1 finding 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 13d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

here; spell the home out. A `rm -rf "$HOME/..."` meant to clear an ephemeral
.claude/skills/kc-entrypoint-change/SKILL.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Editing kube-coder boot scripts safely

Changes to the pod entrypoint affect every workspace on next restart, and a mistake can break boot for everyone. This skill captures the gotchas and the validation loop.

Where boot logic lives

  • charts/workspace/start.sh — the main IDE container entrypoint. Shipped into a ConfigMap via {{ tpl (.Files.Get "start.sh") . | indent 4 }} in templates/workspace-entrypoint-configmap.yaml, and run as command: ["/bin/bash","-l","/workspace-entrypoint/start.sh"].
  • charts/workspace/templates/ssh-server-configmap.yaml — the SSH sidecar entrypoint (data.entrypoint.sh), gated by {{- if .Values.ssh.enabled }}.
  • charts/workspace/templates/terminal-entry-configmap.yaml — the per-ttyd connection script (tmux attach / DEC-mode resets).

Prefer editing start.sh as a real file; the configmaps are YAML block scalars.

Footgun 1 — the two-home layout (MOST IMPORTANT)

There are two home directories:

  • /home/dev — the persistent PVC. Survives restarts.
  • /home/ubuntu — ephemeral. Wiped on every pod restart.

start.sh and the interactive terminals (ttyd, code-server, SSH) run with HOME=/home/ubuntu. That's why /home/ubuntu/.ssh is a symlink onto the PVC. Anything a tool writes under $HOME is lost on restart unless it's redirected onto /home/dev.

But $HOME is not /home/ubuntu everywhere (issue #793). server.py forces HOME=/home/dev on the CLIs it spawns — agent tasks, hooks, the hypervisor — so they find their config on the PVC, and tmux copies the environment of whoever started its server into every session created later. So a bash -l can read either /home/dev/.profile or /home/ubuntu/.profile. Consequences when you touch boot wiring:

  • Seed env blocks into both homes — bootstrap_rc loops over /home/ubuntu /home/dev for exactly this reason. One home only means the block silently does nothing for half the shells in the pod.
  • /home/ubuntu is rebuilt from skel each boot (so it always gets the current block text), while the PVC's /home/dev copy is written once and kept. Changing a block's body needs a new marker string, or the stale text lives on in /home/dev's rc files.
  • Since #795, persist_cred links .ssh, .gitconfig and the per-tool credential paths in both homes at the one store under /home/dev/.credentials/, so ~/.ssh resolves to the same keys regardless of context (anything it displaces is kept at a .pre795.<epoch> backup, not deleted). Still never write ~ or $HOME in docs, examples or scripts here; spell the home out. A rm -rf "$HOME/..." meant to clear an ephemeral copy will delete PVC data when $HOME is /home/dev.

Read the full file on GitHub · 132 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. 13d ago Changed · +23 lines scan A → C 1ae8bcfaf963
  2. 2mo ago First seen · 109 lines · 75 tokens per session scan A 56c7c2853ddd

Subscribe to this mod's changes

kc-entrypoint-change is a skill published in the GitHub repository imran31415/kube-coder (399 stars, last pushed 2d ago), licensed MIT. It adds 75 tokens to every session and 1,722 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

kc-entrypoint-change compared with skills from other repositories

kc-entrypoint-change next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
kc-entrypoint-changeimran31415/kube-coderSkill3992d agoMITC75Claude Code
deploy-docker-composeomnigent-ai/omnigentSkill11kyesterdayApache-2.0A84Claude Code, Codex
bdd-container-updateNVIDIA-AI-Blueprints/video-search-and-summarizationSkill2.0kyesterdayApache-2.0A117Cursor
securing-kubernetes-on-cloudxalgorix/xalgorixSkill1.3kyesterdayApache-2.0A80Claude Code, Codex
detecting-privilege-escalation-in-kubernetes-podsxalgorix/xalgorixSkill1.3kyesterdayApache-2.0A40Claude Code, Codex
docker-socket-mountPurpleAILAB/DecepticonSkill6.5k8d agoApache-2.0B67Claude Code
k8s-networkingrohitg00/kubectl-mcp-serverSkill96815d agoMITA34Claude Code, Codex