aidevtools: Skill for Claude Code

.agents/skills/update-ai-versions/SKILL.md

update-ai-versions is a skill for Claude Code from sugarfreebytes/aidevtools. It costs 29 tokens per session (655 once invoked), scanned A, original, MIT.

A maintenance task that finds the latest published versions of AI coding assistants named in a Dockerfile and updates their version entries in place. A Dockerfile is a recipe for building a software container, an isolated package of an application and its tools.

In plain words
What is it for?
Use it to refresh Claude Code, Codex, Gemini CLI, and any other AI assistant version arguments found in a Dockerfile.
Why use it?
It keeps pinned assistant versions from becoming outdated while leaving unrelated language and tool versions alone. It also uses the appropriate published source for each assistant.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions Claude Code; installed under .agents/ (shared by several agents); mentions Codex.

This is sugarfreebytes/aidevtools's own configuration. It tells Claude Code how to work on aidevtools 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 aidevtools configures →

Reuse

Borrowing it

Nothing to install: this file belongs to sugarfreebytes/aidevtools. 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/sugarfreebytes/aidevtools/main/.agents/skills/update-ai-versions/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/sugarfreebytes/aidevtools

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 update-ai-versions

README.md
[![agentmods](https://agentmods.dev/badge/skills/sugarfreebytes/aidevtools/update-ai-versions/github.svg)](https://agentmods.dev/skills/sugarfreebytes/aidevtools/update-ai-versions)
Your own site
<a href="https://agentmods.dev/skills/sugarfreebytes/aidevtools/update-ai-versions"><img src="https://agentmods.dev/badge/skills/sugarfreebytes/aidevtools/update-ai-versions/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 update-ai-versions

Your own site · 80×15
<a href="https://agentmods.dev/skills/sugarfreebytes/aidevtools/update-ai-versions"><img src="https://agentmods.dev/badge/skills/sugarfreebytes/aidevtools/update-ai-versions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 655 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 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 $0.00029 $0.00655
Opus 5.5 $0.00012 $0.00262
Sonnet 5.5 $0.00006 $0.00131
Haiku 4.5 $0.00003 $0.00065

Measured 1mo ago against content hash 162f0e4b25bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

update-ai-versions scanned grade A 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 1mo 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

allowed-tools: Read, Edit, Bash(npm view *), Bash(curl *), WebFetch, WebSearch
.agents/skills/update-ai-versions/SKILL.md · 38 lines

How it starts

The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Update AI tool versions in the Dockerfile

Your job is to refresh the pinned versions for the AI coding assistants in Dockerfile to whatever is current upstream right now, then edit the Dockerfile in place.

Steps

  1. Read Dockerfile and extract the current values for every ARG *_VERSION line whose tool is an AI assistant. Today that means:

    • CLAUDE_VERSION — Claude Code CLI, installed via https://claude.ai/install.sh
    • CODEX_VERSION — npm package @openai/codex
    • GEMINI_VERSION — npm package @google/gemini-cli

    If new ARG *_VERSION lines have been added for other AI assistants (e.g. opencode, cursor, copilot), include them too. Skip non-AI versions (Java, Node, Python, Go, Deno, NVM).

  2. Look up the latest version of each tool. Prefer the authoritative source per tool:

    • npm packages → npm view <package> version (fast, no auth needed). Both @openai/codex and @google/gemini-cli are on npm.
    • Claude Code CLI → fetch https://claude.ai/install.sh and read the version it defaults to, or check the GitHub releases page for anthropics/claude-code. Pick whichever returns a clean version string.
    • For any other tool: use npm view if it's an npm package, otherwise WebFetch the project's releases page or WebSearch for " latest version".

    Run these lookups in parallel where possible.

  3. Compare each fetched version to what's currently in the Dockerfile. Build a short table of tool: current → latest and print it so the user can see what's about to change.

  4. Edit the Dockerfile: for each tool whose version actually changed, update the corresponding ARG <TOOL>_VERSION=<value> line. Use the Edit tool, one edit per line. Do not touch lines for tools that are already up to date. Do not reformat or reorder anything.

  5. Report the final result: list each tool with unchanged or old → new. Do not commit — leave the diff staged for the user to review.

Notes

  • The Claude install script convention is bash -s <version> with a bare semver (no v prefix). Match the existing format in the Dockerfile for each tool (some have a v prefix, some don't — preserve whatever style is already there).
  • If a lookup fails (network error, package renamed, etc.), report it and skip that tool rather than guessing a version.
  • Do not invent new ARG lines or install new tools — this command only refreshes existing pins.

Read the full file on GitHub · 38 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. 1mo ago First seen · 38 lines · 29 tokens per session scan A 162f0e4b25bc

Subscribe to this mod's changes

update-ai-versions is a skill published in the GitHub repository sugarfreebytes/aidevtools (1 stars, last pushed 1mo ago), licensed MIT. It adds 29 tokens to every session and 655 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-07.

Related

update-ai-versions compared with skills from other repositories

update-ai-versions next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
update-ai-versionssugarfreebytes/aidevtoolsSkill11mo agoMITA29Claude Code
tencentcloud-tkeAceDataCloud/SkillsSkill17yesterdaynoneD94Claude Code, Codex
polar-local-environmentfcakyon/claude-codex-settingsSkill1.2kyesterdayApache-2.0A22Claude Code
kubernetes-operationsWrongStack/WrongStackSkill404yesterdayMITA102Claude Code
neo4j-document-import-skillneo4j-contrib/neo4j-skillsSkill1204d agoMITA187Claude Code
neo4j-aura-agent-skillneo4j-contrib/neo4j-skillsSkill1204d agoMITA158Claude Code
neo4j-modeling-skillneo4j-contrib/neo4j-skillsSkill1204d agoMITA152Claude Code