opencode-cli
Skill0 starsChanged 6 days agoMITView on GitHub
Use when running or scripting the opencode command line or TUI - opencode run in CI, sessions (export, import, continue), models and auth, mcp/plugin/service/serve commands, the shared background server vs --standalone or --server, cli.json settings, keybinds, themes, the web UI, ACP editor integration - or unsure whether an opencode command or flag is v1 or v2.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- amirkiarafiei/subagent-cli-skillstheir own version, first seen Nov 15, 2026
- samirpatil2000/skillstheir own version, first seen Nov 1, 2026
- wentorai/Research-Clawtheir own version, first seen Oct 8, 2026
- spenserhale/skillstheir own version, first seen Sep 24, 2026
- Arcadi4/skillstheir own version, first seen Jul 6, 2026
- julweber/ai_agent_skillstheir own version, first seen Mar 20, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- wentorai/Research-Claw★ 858same name
- samirpatil2000/skills★ 17same name
- amirkiarafiei/subagent-cli-skills★ 14same name
- Arcadi4/skills★ 4same name
- julweber/ai_agent_skills★ 2same name
- Kubementat/ai_agent_skills★ 2same name
- spenserhale/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- capturing-tk-screenshots-headlessSkill
- concept-parableSkill
- opencode-apiSkill
- opencode-buildSkill
- provider-onboardingSkill
- pygameturboSkill
- thonny-pluginSkill
- timestampSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
When they disagree, the higher source wins.
- Very specific: line 52
Stop or start the shared server: opencode service stop|start.
- Very specific: line 59
No cached page for it? The live Markdown of any docs page is its URL with the trailing / replaced by .md.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
opencode CLI docs (v2)
Overview
opencode v2 changed the CLI, and what you remember is probably v1: v2 runs one
shared background server per account (--standalone for a private one,
--server <url> to pick one), sessions live under opencode session ..., and
CLI/TUI settings are in cli.json (not opencode.json). Do not give opencode
commands or flags from memory. Look them up first:
opencode <command> --helpon the installed version, when you can run it.ERRATA.mdbeside this file: known mistakes in the docs below.references/: the v2 CLI docs cached as Markdown.
When they disagree, the higher source wins.
Which reference
| Task | Read |
Starting opencode, automation, shared background service, debug paths, uninstall | references/intro.md |
A command or flag: run, session, auth, models, mcp, plugin, stats, serve, pair, service, api, acp, debug, upgrade | references/commands.md |
| TUI basics: context, shell, side questions, models, history, editor | references/tui.md |
cli.json: appearance, input, sessions, diffs, alerts, terminal, plugins | references/config.md |
| A keybind, or rebinding one | references/keybinds.md |
| Themes: choose, system colors, create one, use one in a plugin | references/theme.md |
| Signing in to providers, credential storage, env vars | references/providers.md |
| Web UI access, ports, password, standalone | references/web.md |
| Editor integration over ACP | references/acp.md |
| Loading TUI-only plugins | references/plugins.md |
commands.md, config.md, keybinds.md and acp.md are 380-630 lines:
list their headings first (grep -n '^##' <file>), then read only the section
you need.
Quick facts (checked against the references)
- CI:
opencode run --standalone --model <provider/model> "<prompt>"with the provider key as a secret env var (commands.md, top). - Export a session redacted:
opencode session export <id> --sanitize. - Stop or start the shared server:
opencode service stop|start.
Keeping it current
- References are verbatim copies; never edit them. Refresh with
python scripts/refresh_references.py(path relative to this skill's folder;--checkreports staleness without writing;--helpfor the rest). - No cached page for it? The live Markdown of any docs page is its URL with
the trailing
/replaced by.md; the index ishttps://opencode.ai/docs/llms.txt. - Cite the page URL without
.md(it is inreferences/manifest.json). - Found a doc mistake (the docs contradict
--helpor real behavior)? Add an entry toERRATA.mdwith your evidence.
Common mistakes
| Mistake | Fix |
opencode export <id> (v1) | opencode session export <id> |
Assuming each TUI or run starts its own server | One shared background server; --standalone for a private one in CI |
Theme/keybinds in opencode.json | cli.json (config.md) |
Adding a TUI-only package to opencode.json | It goes in cli.json; a plugin in opencode.json loads its TUI part automatically (plugins.md) |
| Reading a whole 600-line reference | Search headings, read one section |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.