update-all
Update all coding agent harnesses (Codex, Claude Code, pi, Antigravity CLI) in one run with before/after version reporting.
Skill by cskwork
Chalk Labs
Need help deploying update-all?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/cskwork-update-all
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/update-all
What's inside
- Skill name
update-all- Description
- Update all coding agent harnesses (Codex, Claude Code, pi, Antigravity CLI) in one run with before/after version reporting. Use for /update-all, "update all", "update harnesses", "update codex claude pi agy", "하네스 업데이트".
- License
- MIT
- Instructions
- 82 lines in SKILL.md · view on GitHub
Before installing
- Instructions call the network
Files (1)
- SKILL.md
SKILL.md (first lines)
# update-all Update every installed coding agent harness in one run, then print one before/after version table. | Harness | Binary | Version check | Update command | |---|---|---|---| | Codex | `codex` | `codex --version` | `codex update` | | Claude Code | `claude` | `claude --version` | `claude update` | | pi | `pi` | `pi --version` | `pi update` | | Antigravity CLI | `agy` | `agy --version` | `agy update` | Run every command under the host's ordinary tool execution policy (approvals, sandbox). Do not bypass it. ## 1. Detect and snapshot (before) For each binary, run `command -v <cli>`. - Found: run `<cli> --version`. Record the trimmed output as **Before**. - Version command exits nonzero: record Before as `error (exit N)` and keep its exact stderr for the report. - Not found: mark the harness `not installed`. Skip it in steps 2 and 3. Never guess, infer, or invent a version. ## 2. Update installed harnesses Run the update command for each installed harness, sequentially or in parallel: - `codex update` - `claude update` - `pi update` - `agy update` Use `pi update --all` instead of `pi update` only when the user explicitly asks to include pi extensions or packages. For each command, keep the exact exit code and stderr, plus stdout when it explains the result. - One failure never aborts the run. Continue with the remaining harnesses. - Do not install missing harnesses. - Do not elevate privileges (`sudo`), and do not retry with an alternative installer (npm, brew, curl scripts). ## 3. Verify and report (after) Rerun `<cli> --version` for every harness that was installed in step 1, including harnesses whose update failed. Record **After** the same way as Before. Assign one status per harness: | Status | When | |---|---| | `updated` | Updater exit 0, both version reads succeeded, After differs from Before. | | `unchanged` | Updater exit 0, both version reads succeeded, After equals Before. | | `failed (exit N)` | Updater exited nonzero (N = updater exit code). Stays failed even if After differs from Before. Also used when the updater succeeded but a version read failed (N = that version command's exit code). | | `not installed` | `command -v` found nothing. Nothing was run. | Compare versions only when the updater and both version reads succeeded. No failed command gets a success status. Print exactly one Markdown table, one row per harness, in this shape: ```markdown | Harness | Command | Before | After | Status | |---|---|---|---|---| | Codex | `codex update` | <before> | <after> | updated | | Antigravity CLI | — | — | — | not installed | ``` Use `—` for cells with no value (commands and versions of missing harnesses). Show the command actually run, for example `pi update --all` when the user opted in. Below the table, add one fenced block per failure with the exact stderr. Name the harness and phase (`before --version`, `update`, `after --version`) and the exit code: **pi, update, exit 1** ```text <exact stderr> ``` If stderr is empty, write `(stderr empty)` and include stdout when it explains the failure. ## Done when - Every installed harness had its update attempted, or its failure is reported with exact stderr. - One full before/after table is printed.
From cskwork/skills, MIT.
More from cskwork/skills
- 0.586score
diagram-design
cskwork
Create branded architecture, architecture delta, IT current-state, flowchart, sequence, state machine, ER/data model, timeline, swimlane, quadrant…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
motion-graphics
cskwork
Make code-driven motion-graphics videos rendered from HTML, frame-accurate and reproducible.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
bughunt
cskwork
Find and fix interaction bugs in games, CRUD apps, AI chat and media apps with seeded self-playing personas, per-step invariant oracles, reproducible failure…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
gpt-image-2
cskwork
Generate images with ChatGPT's current image model (GPT Image 2 at the time of writing, and whatever model succeeds it) inside your agent, using your existing…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
image-first-ui
cskwork
Image-first interface creation.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
db-intelligence
cskwork
Unified database intelligence for PostgreSQL, MySQL, SQLite, and MongoDB in one skill: detect the engine, connect credential-safe, extract schema, build an…
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
quality-loop
cskwork
Product quality bar and improvement loop for iPhone apps, websites, web apps, games, and other software.
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
sdlc-kit
cskwork
Four-phase SDLC loop for any codebase: intent (90% clear: explore, reproduce, research, ask the requester), plan (parallel scouts, one PLAN.md), build…
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
mobile-ui-ux
cskwork
Measurable usability and interaction rules for iPhone apps, mobile web, and mobile games, with a 7-category scoring rubric, an independent-judge prompt, and a…
SkillAFreshness: A★ 3Claude Code +9 - 0.580score
jk
cskwork
Manage Jenkins controllers with jk, including jobs, runs, logs, artifacts, credentials, nodes, queues, and plugins.
SkillAFreshness: A★ 3Claude Code +9 - 0.578score
sol
cskwork
Orchestrate application work through GPT-6.1 Sol subagents (high reasoning, fast tier).
SkillAFreshness: A★ 3Claude Code +9 - 0.575score
call-agent
cskwork
Delegate a bounded task to a user-named peer CLI, or to fill a capability the host lacks.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Oct 24, 2026 (7 days ago)
- Last release
- Oct 24, 2026 (7 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/cskwork/skills
- Website
- cskwork.github.io/skills/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from cskwork
- 0.586score
bughunt
cskwork
Find and fix interaction bugs in games, CRUD apps, AI chat and media apps with seeded self-playing personas, per-step invariant oracles, reproducible failure…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
image-first-ui
cskwork
Image-first interface creation.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
motion-graphics
cskwork
Make code-driven motion-graphics videos rendered from HTML, frame-accurate and reproducible.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
gpt-image-2
cskwork
Generate images with ChatGPT's current image model (GPT Image 2 at the time of writing, and whatever model succeeds it) inside your agent, using your existing…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
diagram-design
cskwork
Create branded architecture, architecture delta, IT current-state, flowchart, sequence, state machine, ER/data model, timeline, swimlane, quadrant…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
db-intelligence
cskwork
Unified database intelligence for PostgreSQL, MySQL, SQLite, and MongoDB in one skill: detect the engine, connect credential-safe, extract schema, build an…
SkillAFreshness: A★ 3Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get update-all running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
update-all skill. Agent Mart. https://agentmart.dev/skills/cskwork-update-all (last updated 2026-10-24)