Paste this into your coding agent with your project open. It checks your installation and installs the updater for you.
Upgrade my existing Opchain installation to the latest published release, including oc-update. I’m on 1.9 or earlier, so that command may not exist yet. Check this project’s installation and my agent host first. If Opchain is managed as a global plugin, use the host’s supported plugin update flow. Do not create a duplicate repo installation or overwrite plugin caches. If this is an Opchain source checkout, explain that syncing local files does not fetch a public release and confirm the source update path with me. For a repo installation, verify Node.js 22.13 or newer and check https://opchain.dev/opchain-update/latest.json. If 2.0 or newer is not published, stop and tell me. Otherwise, download https://opchain.dev/update.mjs completely to a temporary file. Run it from my project root with --check first, then run the update for the detected installation. Do not copy SKILL.md files by themselves. Preserve my checkpoints, tracking consent, usage history and unrelated skills. Use the updater’s backup of replaced Opchain files and report its location. Keep learning and hooks at their current settings. Verify the installed version and complete files afterward. Report any failure or tracking warning accurately. Tell me if I need a new agent session, then confirm that /oc-update check works. Do not commit or deploy my project.
Each skill includes a SKILL.md file and its supporting files. Install complete folders so references and runtime commands work. The local runtime needs Node.js 22.13 or newer.
Need the CLI first?Install Claude Code, then pick a path. The plugin is the one to use if you can — it carries the skills and the hooks that enforce them. The zip carries the skills only.
Two commands. Adds the 36 skills, sixteen registered slash commands, and three hooks: a commit gate that refuses git commit unless the code being committed passed bug-check, pipeline state at session start, and a pointer to the next skill when one finishes.
/plugin marketplace add asfbay-bit/opchain-skills
/plugin install opchainDrop the skills into .claude/skills/; Claude Code discovers them next session. Use this where a plugin install isn't an option. Note what it does not include: none of the hooks, so nothing mechanically enforces a gate — the skills describe the contract and you follow it.
curl -L https://opchain.dev/opchain-skills.zip -o opchain-skills.zip
unzip opchain-skills.zip -d .claude/skills/
claude
> /oc-discoverVerify it worked. In that session, type/oc-discover. If oc-app-architect starts an interview ("Who's the primary user? Sync or async? …"), you're set. If not, re-check that the files landed in.claude/skills/<name>/SKILL.md.
/oc-discover). Naming it explicitly is markedly more reliable than hoping the description matches.Commit the skills directory so every teammate uses the same prompt pipeline.
git add .claude/skills/
git commit -m "chore: add opchain skills"
git pushCodex — and any MCP-aware agent (Claude Desktop, Cursor, Windsurf) — reaches opchain two ways, over theModel Context Protocolor as drop-in skills. The drop-in path is the closest match to Claude Code: Codex Agent Skills auto-trigger by description, so the pipeline fires on its own. The hosted MCP endpoint needs no install, but relies on the model choosing to call opchain's tools.
Drop the skills in (recommended) — Codex Agent Skills use the same SKILL.md format and auto-trigger by description — the same unenforced, name-it-yourself matching Claude Code skills rely on:
curl -L https://opchain.dev/opchain-skills.zip -o opchain-skills.zip
unzip opchain-skills.zip -d .agents/skills/
codexOr point Codex at the hosted MCP server — nothing to install; works for any MCP client:
# ~/.codex/config.toml
[mcp_servers.opchain]
url = "https://opchain.dev/mcp"Verify it worked. Run codexand ask "build me an app" — a drop-in skill match (or, over the hosted server, opchain's route tool) hands you to oc-app-architect's discovery interview. Note: Codex doesn't surface MCP prompts, so opchain's /oc-* verbs aren't slash commands in Codex — just describe what you want ("build me an app", "audit this") and the matching skill triggers.
Already running opchain in a project? Replace the skills in place. Your .checkpoints/ directory livesoutside .claude/skills/, so every skill's session state — phase, step, next actions, decisions — is preserved across the swap. The new skills pick up the existing checkpoints on their next invocation.
/oc-update check
/oc-updateIf 1.x does not recognize /oc-update, use the bootstrap installer below once. Run it from your project root with Node.js 22.13 or newer. For Codex, use --target=codex.
Run from your project root. The installer checks the bundle, backs up replaced Opchain files and installs the current release.
installer=$(mktemp -t opchain-update)
curl --fail --location https://opchain.dev/update.mjs -o "$installer" &&
node "$installer" --target=claude
rm -f "$installer"Verify the installation. Start a new agent session and run/oc-update check. It should report the installed release as current. Your saved checkpoints remain available.
If you committed .claude/skills/ with the team flow, do the upgrade on a branch so the skill diff lands in review:
git switch -c chore/upgrade-opchain
installer=$(mktemp -t opchain-update)
curl --fail --location https://opchain.dev/update.mjs -o "$installer" &&
node "$installer" --target=claude
rm -f "$installer"
git diff --statReview the changed Opchain files and install receipt before committing the update.
/oc-update checkChecks the installed skills against the current release without changing files.
Skills don't phone home, so a stale install stays stale silently. This Claude Code hook checks at session start: it reads theversion from your installed skills' frontmatter, compares it against the live catalog'scatalogVersion atopchain.dev/skills.json, and drops a one-line notice into the session when a newer release is out. Merge it into ~/.claude/settings.json (or a project's .claude/settings.json) — if you already have a hooks block, add theSessionStart entry to it rather than pasting the whole object.
{
"hooks": {
"SessionStart": [
{
"matcher": "startup",
"hooks": [
{
"type": "command",
"command": "f=$(ls \"${CLAUDE_PROJECT_DIR:-.}\"/.claude/skills/oc-*/SKILL.md \"$HOME\"/.claude/skills/oc-*/SKILL.md 2>/dev/null | head -n1); [ -n \"$f\" ] || exit 0; installed=$(sed -n 's/^version:[[:space:]]*//p' \"$f\" | head -n1); [ -n \"$installed\" ] || exit 0; latest=$(curl -sf --max-time 3 \"https://opchain.dev/skills.json?u=$(date +%s)\" | sed -n 's/.*\"catalogVersion\": *\"\\([^\"]*\\)\".*/\\1/p' | head -n1); [ -n \"$latest\" ] && [ \"$installed\" != \"$latest\" ] && echo \"opchain skills v$latest is available (installed: v$installed). Update guide: https://opchain.dev/install#update\"; exit 0"
}
]
}
]
}
}It checks, it never installs — releases can carry breaking changes (see the changelog), so the swap stays a deliberate step: run the upgrade command above when you see the notice. When you're current, offline, or opchain isn't installed, the hook prints nothing and gets out of the way. It fires only onstartup — not on resume or compact — and gives the network three seconds before giving up.
.claude/skills/.checkpoints/orchestrator.md.agents/skills/~/.codex/config.toml[mcp_servers.opchain]..checkpoints/; pick up exactly where you left off, next chat or next weekWant a preview first? The demo walks you through fifteen end-to-end opchain examples.
No code yet · greenfield project
/oc-discover — oc-app-architect runs the spec interviewCode already on disk · no opchain history
/oc-rev-spec — oc-reverse-spec backfills a spec from the code/oc-ops for the orchestrator's read on what's openBug or feature already in progress