AGENT SKILL

workspace-orchestration

Use when working in a persistent-slot workspace that uses docko. Covers slot claims, delegated teammate inheritance, and the rule that code edits inside slots/ require an active claim.

License 许可
MIT
最近更新
2026-11-14
Stars
★ 3.3w
Forks
3.9k

01 前言 Front Matter

name
workspace-orchestration
description
Use when working in a persistent-slot workspace that uses docko. Covers slot claims, delegated teammate inheritance, and the rule that code edits inside slots/ require an active claim.

02 资源文件 8

03 SKILL.md 正文

docko Workspace Orchestration

Use this skill whenever work touches files in slots/ or when the user asks about workspace coordination.

Quick Path

  1. Run /dock-status or docko status --brief. docko walks up to the workspace root on its own, so this works from inside a slot; pass --root <workspace> only from outside the workspace.
  2. Pass --session "$DOCKO_SESSION_ID" on every claim, heartbeat, and release. The SessionStart hook exports it for this session.
  3. Prefer docko slot acquire --session "$DOCKO_SESSION_ID" --branch <branch> --task "<task>" --brief when you want docko to pick a slot. Selection is round-robin, starting after the last slot claimed for that application.
  4. In a workspace with several applications, pass --application <id> so the pick stays inside the right pool.
  5. Add --prefer <slot-id> when a specific slot is the right one (it already has the branch, say). docko takes it when free and rotates normally when it is not.
  6. If every slot is busy and docko asks whether it should create a fresh managed clone, answer explicitly.
  7. Use docko claim --session "$DOCKO_SESSION_ID" --resource slot --id <slot> --branch <branch> --task "<task>" only when you already know the exact slot.
  8. Do the work inside that slot. On long-running work, keep the claim fresh: docko heartbeat --session "$DOCKO_SESSION_ID" --resource slot --id <slot>
  9. Release the slot when done: docko release --session "$DOCKO_SESSION_ID" --resource slot --id <slot>

Prefer slash commands when installed:

  • /dock-status
  • /dock-claim <slot> <branch> <task>
  • /dock-heartbeat <slot>
  • /dock-release <slot>
  • /dock-doctor

Rules

  • Work from the root. Edit code in slots/*.
  • If $DOCKO_SESSION_ID is empty (plugin hooks not installed, or a different runtime), stop before the first claim and resolve a real session id — never run claim or heartbeat with an empty --session. Use docko session list --brief or docko session current to pick one.
  • Never invent a session id. The PreToolUse hook checks the runtime's own session, so a made-up id claims a slot that then blocks your own writes. Use $DOCKO_SESSION_ID, or an id from docko session list --brief.
  • branch is claim metadata. docko records it and never runs git checkout.
  • Claims are slot-scoped. Two sessions cannot share a slot, and one claim does not reserve a branch, a PR, or a file.
  • If a parent session already owns the slot, reuse that authority instead of creating a second claim.
  • Subagents started with the Agent tool share the parent's session id, so they inherit the parent's claim and need no docko call. A separately launched claude process gets its own session id and must be delegated: docko delegate --session <owner> --child-session <child> --resource slot --id <slot>.
  • If docko reports AMBIGUOUS_SESSION, run the suggested_command from the error payload, or retry with an explicit --session <id> from docko session list --brief; do not end existing sessions unless the user asked for cleanup.
  • If a write is blocked, read the deny message: it names the slot, the reason (slot-not-claimed, claim-expired, unrelated-session) and the exact command that fixes it.
  • If every slot is busy and the user already approved the fallback, add --clone-when-busy to docko slot acquire.
  • If docko is not runnable, check DOCKO_BIN. If it still fails, stop and tell the user.
  • Do not inspect slots one by one or replace the CLI with docko/registry.json during normal work. Use docko status --brief --claimed instead.

04 同仓库更多 Skills

davila7
davila7/claude-code-templates

wealthville-pool-scoring

Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter/Hold/Exit/Reduce/Avoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

technical-debt-auditor

Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO/FIXME/HACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like "where should I focus refactoring", "find code smells", "audit my codebase", "what technical debt do we have", "prioritize cleanup work", or asks about code quality metrics. Also trigger when the user mentions "code rot", "legacy cleanup", "refactoring priorities", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say "technical debt" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

readonly-sql-guard

Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production/ERP/analytics database (a "query" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write/DDL/side-effect/transaction-escape risks. Checks what a statement CALLS, not just its shape.

数据分析师后端与接口Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

web-artifacts-builder

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

前端开发
★ 3.3w
davila7
davila7/claude-code-templates

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

测试与质量前端开发
★ 3.3w
davila7
davila7/claude-code-templates

discernment-nudge

After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.

内容创作者写作与文案Skill与智能体开发
★ 3.3w