update

A tool for reviewing an existing project's Claude Code setup against current recommended practices.

In plain words
What is it for?
Use it to inspect Claude tooling, choose specific upgrades, and apply approved changes while deciding whether each file should be merged or replaced.
Why use it?
It helps find outdated or missing setup without requiring you to compare the project manually with the latest documentation.

Skill for Claude CodeCodex

Part of the onboard plugin — 8 skills, 5 agents shipped together

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.

agentmods
npx agentmods add skills/apurvbazari/claude-plugins/update
Any agent
npx skills add ApurvBazari/claude-plugins --skill update
Clone the repo
git clone --depth 1 https://github.com/ApurvBazari/claude-plugins

Made for: Claude Code, Codex.

Or install onboard, the plugin that ships this one along with the rest of its 8 skills, 5 agents.

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 8,738 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00054 $0.08738
Opus 5 $0.00027 $0.04369
Sonnet 5 $0.00011 $0.01748
Haiku 4.5 $0.00005 $0.00874

Measured 2d ago against content hash 2b656281305d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

update 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 2d 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.

onboard/skills/update/SKILL.md · 455 lines

How it starts

The opening of the file, as written. The whole thing — 455 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Update Skill — Evolve Tooling to Latest Best Practices

You are running the onboard update skill. This checks whether the project's Claude tooling is aligned with the latest best practices and offers targeted upgrades.

This is NOT a snapshot diff. It's a forward-looking check against current best practices.


Step 0: Initialize Phase Tracking

Runs first, before the Prerequisites Check. This wires the durable phase-task list that the rest of the flow transitions for in-session visibility.

First, read the contract: ../start/references/phase-tracking.md (the cross-skill source of truth — start is its worked example). It defines the task subjects, the pending → in_progress → completed (plus deleted) state machine, and the HARD GATE mapping. Follow it verbatim — do not re-derive its rules here. update runs on an already-onboarded project (an onboard-meta.json exists by the time the baseline check in Phase 0 passes), so it has no cross-session resume probe of its own; the task list here is in-session progress visibility, and the durable record lives in the on-disk artifacts the contract describes.

Create the 8 update phase tasks via TaskCreate, all status: "pending", one per ladder phase, using the update: subject prefix from the contract's § Other entry points table:

Subject activeForm
update:verify-baseline Verifying the onboard baseline
update:read-artifacts Reading existing artifacts
update:reanalyze Re-analyzing the codebase
update:best-practices-drift Checking best practices & detecting drift
update:present-preview Presenting findings & rendering preview
update:approve-gate Awaiting approval of upgrades
update:apply Applying approved updates
update:summary Summarizing changes

Only this orchestrator touches the list. Every internal skill it later invokes (onboard:generate, onboard:research) is task-blind per the contract's § Ownership — when update invokes generate inside Phase 6, that does not create its own list; it runs inside this Phase-6 task. Routing note (per the contract's § Routing): if Phase 0 routes into adopt (missing-baseline branch), adopt owns its own adopt: list and marks it complete on handoff; the two lists coexist — this update list never touches adopt's tasks and vice versa.

Read the full file on GitHub · 455 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 455 lines · 54 tokens per session scan A 2b656281305d

Subscribe to this mod's changes

update is a skill published in the GitHub repository ApurvBazari/claude-plugins (0 stars, last pushed 26d ago), licensed MIT. It adds 54 tokens to every session and 8,738 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.