bb-subagents
Launch a new bb worker thread with an explicit project, provider, model, reasoning effort, fast/default tier, worktree, and prompt. Use when the user says "launch a bb session", "bb subagent", "spawn a bb thread", "new worktree agent", or asks to run Grok/Codex/Cursor CLI in bb on a repo. Differentiator vs bb-cli: this is only the spawn recipe. Differentiator vs codex-subagent/launch-subagent: those are Codex CLI and Cursor Task tool, not bb threads.
davidondrej · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add davidondrej/skills --skill bb-subagents技能说明
name: bb-subagents description: 'Launch a new bb worker thread with an explicit project, provider, model, reasoning effort, fast/default tier, worktree, and prompt. Use when the user says "launch a bb session", "bb subagent", "spawn a bb thread", "new worktree agent", or asks to run Grok/Codex/Cursor CLI in bb on a repo. Differentiator vs bb-cli: this is only the spawn recipe. Differentiator vs codex-subagent/launch-subagent: those are Codex CLI and Cursor Task tool, not bb threads.'
bb-subagents
Spawn a bb thread the way the user launches them. Read this before any bb thread spawn.
Use bb-cli for status, tell, wait, inspect, automations. This skill is spawn-only.
Default worker: Codex gpt-5.6-sol, --reasoning-level high, --service-tier fast ("GPT 5.6 Sol High Fast"); bump to xhigh for larger refactors. Use this unless the user requests otherwise.
Hard rules
- Pass
--projectexplicitly. The CLI does not infer it from the current thread. - Never guess provider or model IDs. Lookup, then spawn.
- After spawn: do not
bb thread open,--split, or focus the new thread unless the user asks. Just launch. Report the thread id. Stop. - Do not poll,
bb thread wait, or read logs unless the user asks. - Prefer bare
bb. IfBB_CLIis set,"$BB_CLI"is fine.
Recipe
bb project list --json
bb provider list --json
bb provider models <provider-id> --json
bb thread spawn --json \
--project <project-id> \
--new-environment worktree \
--provider <provider-id> \
--model <model-id> \
--reasoning-level <level> \
--permission-mode full \
--title "Short title" \
--prompt "..."
Add --service-tier fast only when the user says fast.
Parent / sidebar nest
--parent-self and --parent-thread are optional. Use them when it makes sense — not on every spawn.
--parent-self parents the new thread to the current thread (BB_THREAD_ID). The left sidebar then nests the child under this thread (indented, expandable), like a worker under a manager.
--parent-thread <id> parents it to a specific other thread. Do not combine it with --parent-self.
Omit both for a root thread — its own top-level sidebar row, not nested.
When it makes sense: this thread is coordinating the work, the user asked for a worker under this session, or the job is a clear subtask of this thread. Skip it for unrelated one-off work that should sit as its own top-level thread.
# add when it makes sense — not required
--parent-self
environmentId is often null in the spawn JSON. The worktree is still creating. That is fine.
Lookup
Project. Match name (e.g. DeepAPI) in bb project list --json. Use that id. Confirm sources[].path is the repo the user named. Do not hardcode project IDs.
Provider. From bb provider list --json:
| User says | Provider id |
|---|---|
| Cursor CLI / Cursor subscription | acp-cursor |
| Codex / Codex subscription / ChatGPT sub | codex |
| Claude Code | claude-code |
Do not use acp-grok for "Grok via Cursor CLI". That is Grok Build, a different product.
Model + reasoning. From bb provider models <id> --json. Use the catalog id. Match the user's words to displayName and supportedReasoningEfforts[].description. Those two are not the same field.
If the user does not name a reasoning level, use that model's defaultReasoningEffort from the catalog. Do not pick extra high or max on your own. Example: Claude Code Fable 5 defaults to high.
Known mappings (re-fetch if spawn rejects them)
Cursor CLI — Grok 4.6 Extra High Fast
User: "Grok 4.6 extra high fast (via Cursor CLI subscription)"
--provider acp-cursor \
--model cursor-grok-4.6-medium \
--reasoning-level xhigh \
--service-tier fast
Traps:
- The model id stays
cursor-grok-4.6-mediumeven for Extra High. Extra High is a reasoning value, not a different model id. - On this catalog,
highdisplays as Fast.xhighdisplays as Extra High. - "Extra High Fast" =
xhighplus--service-tier fast. Fast-the-reasoning-label and Fast-the-service-tier are different flags. - Cursor ACP has no
autopermission mode. Usefulloraccept-edits. - Do not pass Cursor Task-tool slugs like
cursor-grok-4.6-high-fast. Those are not bb model ids.
Codex — GPT 5.6 Sol Max
User: "GPT 5.6 Sol Max, via the Codex subscription"
--provider codex \
--model gpt-5.6-sol \
--reasoning-level max
Traps:
- Use native
gpt-5.6-sol, notcursor/gpt-5.6-sol(opencodex routed through Cursor). - Max is
--reasoning-level max, not a model id suffix. - Codex supports
autopermission mode. Still usefullwhen the worker must hit prod, network, or leave the sandbox. - Pass
--service-tier fastonly if the user asks for fast on Codex too.
Worktree
--new-environment worktree is the bb-managed worktree. Do not create a git worktree by hand unless the user says so.
- New worktrees get tracked files only.
- Untracked files (
.env) copy only if the repo has.worktreeinclude. - Optional:
--base-branch <branch>. Omit to use the project's default. - Do not combine
--machinewith an existing--environmentid.
For repos that must keep .env in worktrees, the repo needs .worktreeinclude plus optional .bb-env-setup.sh. Details: bb guide environments.
Prompt
The worker starts blind. It does not see this chat. Write the full brief into --prompt.
Include:
- Objective — one sentence.
- Constraints — read-only vs implement; no git push; no prod writes.
- Skills / files to use — name them. Skills do not carry over. Example: prod reads →
~/.claude/skills/read-prod-database/SKILL.md. - Deliverable — what to return.
- Validation — how it knows it is done.
- Report back — concise report only, or diff + files changed.
Read-only jobs must say no code changes, no git writes, no database writes in the prompt, even if permi
阅读完整来源 ↗