using-hotl
Use when the user explicitly asks for HOTL routing or needs guidance choosing the right HOTL workflow skill.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
01 前言 Front Matter
- name
- using-hotl
- description
- Use when the user explicitly asks for HOTL routing or needs guidance choosing the right HOTL workflow skill.
02 资源文件 1
- 参考文档agents/openai.yaml266 B
03 SKILL.md 正文
When to Use HOTL Skills
HOTL skills are for code-changing tasks that require planning — new features, refactors, and significant changes. Not every task needs a skill.
Answer directly without invoking a skill:
- Code understanding questions — "how does this work?", "where is this defined?", "explain this error"
- Quick fixes — typos, config values, import paths, obvious one-line corrections
- Error tracing — "what does this error mean?", "which file throws this?"
Use hotl:systematic-debugging (no brainstorm/plan needed):
- Bug investigations, test failures, unexpected behavior
Use the full HOTL workflow (brainstorm → plan → execute):
- New features, significant refactors, architectural changes
Available HOTL Skills
Invoke any of these when appropriate:
| Skill | When to Use |
|---|---|
hotl:brainstorming |
Before any feature work — design with HOTL contracts |
hotl:writing-plans |
After design approval — produces docs/plans/YYYY-MM-DD-<slug>-workflow.md |
hotl:governed-execution |
Preferred execution entry point — selects host driver/profile while preserving renewable ownership, bounded loops, effect evidence, and receipts |
hotl:executing-plans |
Linear execution with human checkpoints |
hotl:loop-execution |
Execute a workflow file with loops + auto-approve |
hotl:subagent-execution |
Delegated step runner over the loop execution engine — delegates eligible steps to fresh subagents |
hotl:dispatch-agents |
2+ independent tasks that can run in parallel |
hotl:finishing-a-development-branch |
After execution — merge back, publish/PR, keep, or discard the execution branch/worktree |
hotl:tdd |
Before writing any implementation code |
hotl:systematic-debugging |
When encountering any bug or unexpected behavior |
hotl:skill-authoring |
When creating, editing, or reviewing HOTL skills, agents, commands, or behavior-shaping instructions |
hotl:document-review |
Optional — review existing docs, external specs, or hand-authored plans |
hotl:requesting-code-review |
Dispatched by executors at review checkpoints — standardizes what context the reviewer receives |
hotl:receiving-code-review |
Invoked when review findings arrive — verify, evaluate against contracts, then implement |
hotl:code-review |
After completing implementation, before merging |
hotl:pr-reviewing |
Review a PR across multiple dimensions — description, code, scan, tests |
hotl:resuming |
Resume an interrupted workflow run with verify-first effect recovery and explicit ownership takeover/handoff |
hotl:verification-before-completion |
Before claiming work is done |
hotl:setup-project |
To generate adapter files for Codex, Cline, Cursor, Copilot |
Red Flags (You Are Over-Routing)
- Routing a code question through brainstorming → just answer it
- Creating a plan for a typo fix → just fix it and verify
- Brainstorming before debugging → use systematic-debugging directly
- Skipping brainstorming for a real feature → invoke the skill
HOTL Operating Principles
Human-on-the-Loop: Set intent + constraints upfront. AI executes autonomously within guardrails. Human reviews final output.
Long-running harness contract: Host goals, automations, background agents, hooks, and handoffs provide scheduling and liveness only. HOTL's renewable controller, serialized state, runtime loop/budget stops, effect reconciliation, explicit finish disposition, and sufficient receipt remain authoritative.
Three contracts every implementation workflow should define:
- Intent contract: objective, constraints, success criteria
- Verification contract: how to confirm each step worked
- Governance contract: approval gates, risk level, rollback strategy
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.