auto
Run completely AUTONOMOUSLY — loop the FULL task engine over the Beads queue, a fresh `claude -p` per task, until the queue is…
@23ag1Install SkillTesting
Skill
/completely
Overview and entry point for the completely harness — a quality-first agent workflow unifying GSD (planning depth), Ralph (autonomous loop), and Beads (the spine), under deterministic gate hooks and a default-FAIL evaluator. Use to see what's installed, the command surface, and
Claude Code
$ npx -y skills add 23ag1/completely --skill completely --agent claude-code
Codex
$ npx skills add 23ag1/completely --skill completely -a codex -y
Cursor
$ npx skills add 23ag1/completely --skill completely -a cursor -y
Gemini CLI
$ npx skills add 23ag1/completely --skill completely -a gemini-cli -y
GitHub Copilot
$ npx skills add 23ag1/completely --skill completely -a github-copilot -y
OpenCode
$ npx skills add 23ag1/completely --skill completely -a opencode -y
Windsurf
$ npx skills add 23ag1/completely --skill completely -a windsurf -y
SKILL.md
completely.SKILL.mdname: completely
description: Overview and entry point for the completely harness — a quality-first agent workflow unifying GSD (planning depth), Ralph (autonomous loop), and Beads (the spine), under deterministic gate hooks and a default-FAIL evaluator. Use to see what's installed, the command surface, and how to start.
version: 0.2.0
user-invocable: true
**completely** makes it hard for an agent to quietly cut a corner, fake "done", or ship a stub. Short CLI: `cmpl`. Slash: `/completely:*`.
Command surface
- `/completely:init` — scaffold the thin layer (DoD, CLAUDE.md rules, project gate) into this repo.
- `/completely:sync` (`cmpl sync`) — migrate markdown task state into Beads, idempotently.
- `/completely:run` — drive `bd ready`: supervised (GSD wave subagents) or unattended (Ralph loop).
- `/completely:doctor` (`cmpl doctor`) — upstream version drift + overlay health.
How it fits (one engine, one spine)
Beads = the spine (status + memory via comments/notes/remember + coordination via swarm/gate/merge-slot). GSD = planning depth (discuss→plan→plan-checker loop). Ralph = the OS-level autonomous loop. completely = the quality floor (gate hooks + default-FAIL `evaluator` + no-stub contract) under all of them. GSD and Ralph are two autonomy *modes* of the same loop over `bd ready`, not a choice of tools.
See `plugin/core/HARNESS.md` (the principle), `plugin/core/roles.md` (who owns what), and `docs/TOOL-COMPATIBILITY.md` (the full design + per-tool detail).
Read more
name: completely description: Overview and entry point for the completely harness — a quality-first agent workflow unifying GSD (planning depth), Ralph (autonomous loop), and Beads (the spine), under deterministic gate hooks and a default-FAIL evaluator. Use to see what's installed, the command surface, and how to start. version: 0.2.0 user-invocable: true
**completely** makes it hard for an agent to quietly cut a corner, fake "done", or ship a stub. Short CLI: `cmpl`. Slash: `/completely:*`.
Command surface
- `/completely:init` — scaffold the thin layer (DoD, CLAUDE.md rules, project gate) into this repo.
- `/completely:sync` (`cmpl sync`) — migrate markdown task state into Beads, idempotently.
- `/completely:run` — drive `bd ready`: supervised (GSD wave subagents) or unattended (Ralph loop).
- `/completely:doctor` (`cmpl doctor`) — upstream version drift + overlay health.
How it fits (one engine, one spine)
Beads = the spine (status + memory via comments/notes/remember + coordination via swarm/gate/merge-slot). GSD = planning depth (discuss→plan→plan-checker loop). Ralph = the OS-level autonomous loop. completely = the quality floor (gate hooks + default-FAIL `evaluator` + no-stub contract) under all of them. GSD and Ralph are two autonomy *modes* of the same loop over `bd ready`, not a choice of tools.
See `plugin/core/HARNESS.md` (the principle), `plugin/core/roles.md` (who owns what), and `docs/TOOL-COMPATIBILITY.md` (the full design + per-tool detail).
Stats
6
Stars
0
Forks
Maintained
Maintenance
Shell
Language
MIT
License
Last commit
Created
Ships withcompletely
A quality-first harness for autonomous AI coding agents. It turns "the agent said it's done" into *"here's the proof — graded by an independent, default-FAIL checker."* Done is earned, not asserted.
616 skills9 agents
Stars on GitHub
Momentum
- stars in total
- 6
Other skills on completely.
@23ag1Install Skillcheck
Run all configured quality checks (lint, types, tests) in one pass with terse output — reports "clean" when green, and only the…
@23ag1Install Skillcontrol
Run completely UNDER CONTROL — execute the SINGLE next Beads task through the FULL task engine (understand → map → plan-check →…
@23ag1Install Skillinit
Scaffold the completely thin layer into the current repository — Definition of Done, the harness rules snippet in CLAUDE.md, and…
@23ag1Install Skillplan
Plan a phase or feature DIRECTLY into Beads — no PLAN.md, no markdown bridge. Runs GSD-style socratic discovery + decomposition +…
@23ag1Install SkillMore skills from other plugins.
api-review
Produce the "Playwright 1.xx API Review" artifact for a range of commits - the public API additions and the breaking or behavior…
@microsoftInstall Skill
promptfoo-evals
Write, refine, run, and QA promptfoo evaluation suites: promptfooconfig.yaml, prompts, providers, vars, tests, assertions,…
@promptfooInstall Skilldeepeval
DeepEval evaluation workflow for AI agents and LLM applications. TRIGGER when the user wants to evaluate or improve an AI agent,…
@confident-aiInstall SkillQuestions about completely.
- Which coding agents can use completely?
- completely installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.