惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

博客园 - 三生石上(FineUI控件)
博客园 - Franky
GbyAI
GbyAI
B
Blog
WordPress大学
WordPress大学
D
Docker
小众软件
小众软件
月光博客
月光博客
博客园 - 【当耐特】
T
The Blog of Author Tim Ferriss
IT之家
IT之家
腾讯CDC
Engineering at Meta
Engineering at Meta
Vercel News
Vercel News
H
Help Net Security
M
MIT News - Artificial intelligence
L
LangChain Blog
云风的 BLOG
云风的 BLOG
S
SegmentFault 最新的问题
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
美团技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
V
V2EX
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻

Recent Commits to openclaw:main

test: merge chat side-result checks · openclaw/openclaw@ddd2c2a test: merge cron history checks · openclaw/openclaw@f7eb746 test: merge responsive navigation shell checks · openclaw/openclaw@c2e4b47 docs(changelog): add codex oauth fixes · openclaw/openclaw@628e6cd test: merge navigation routing cases · openclaw/openclaw@5d8cecb Tests: mock channel registry bundled fallback · openclaw/openclaw@2b08233 Secrets: avoid broad web search discovery for single plugin config · openclaw/openclaw@a464f59 test: merge config view browser checks · openclaw/openclaw@20cf511 fix(status): align oauth health with runtime · openclaw/openclaw@eed7116 feat: add macOS screen snapshots for monitor preview (#67954) thanks … · openclaw/openclaw@f377db1 fix: report shared auth scopes in hello-ok (#67810) thanks @BunsDev · openclaw/openclaw@0b6c39b Auto-reply: avoid eager bundled route fallback · openclaw/openclaw@3ea1bf4 Tests: narrow session binding contract setup · openclaw/openclaw@54e4e16 fix(macOS): enable undo/redo in webchat composer text input (#34962) · openclaw/openclaw@00951dc Tests: speed up channel setup promotion · openclaw/openclaw@82b529a Docs: refresh agent instructions · openclaw/openclaw@5775fe2 fix(auth): serialize OAuth refresh across agents to fix #26322 (#67876) · openclaw/openclaw@8e79080 test: allow ollama public surface boundary test · openclaw/openclaw@7d4f1a6 Docs: add test performance guardrails · openclaw/openclaw@89706d3 Tests: restore context-engine usage proof · openclaw/openclaw@e4c4f95 Tests: slim context engine runtime coverage · openclaw/openclaw@74c198f ci: retry failed custom checkouts · openclaw/openclaw@0ee5baf test: trim duplicate provider auth onboarding cases · openclaw/openclaw@1ffc02e matrix: fix sessions_spawn --thread subagent session spawning (#67643) · openclaw/openclaw@1ce2596 test: reduce auth choice fixture churn · openclaw/openclaw@857b9cd test: mock health status config boundaries · openclaw/openclaw@9d5ab4a test: mock onboard config io boundary · openclaw/openclaw@299694d test: mock legacy state plugin boundaries · openclaw/openclaw@2713089 test: mock channel install boundaries · openclaw/openclaw@b945248 test: mock doctor preview channel boundaries · openclaw/openclaw@b1a3ad4
docs: add steer command guide · openclaw/openclaw@90a5b08
steipete · 2026-05-04 · via Recent Commits to openclaw:main

@@ -0,0 +1,78 @@

1+

---

2+

summary: "Steer an active run without changing queue mode"

3+

read_when:

4+

- Using /steer or /tell while an agent is already running

5+

- Comparing /steer with /queue steer

6+

- Deciding whether to steer the current run, a sub-agent, or an ACP session

7+

title: "Steer"

8+

sidebarTitle: "Steer"

9+

---

10+11+

`/steer` sends guidance to an already-active run. It is for "adjust this

12+

run while it is still working" moments, not for starting a new turn.

13+14+

## Current session

15+16+

Use top-level `/steer` to target the active run for the current session:

17+18+

```text

19+

/steer prefer the smaller patch and keep the tests focused

20+

/tell summarize before making the next tool call

21+

```

22+23+

Behavior:

24+25+

- Targets only the current session's active run.

26+

- Works independently of the session's `/queue` mode.

27+

- Does not start a new run when the session is idle.

28+

- Replies with a warning when there is no active run to steer.

29+

- Uses the active runtime's steering path, so the model sees the guidance at

30+

the next supported runtime boundary.

31+32+

## Steer vs queue

33+34+

`/queue steer` changes how normal inbound messages behave when they arrive

35+

while a run is active. `/steer <message>` is an explicit command that tries to

36+

inject that command's message into the active run at the next supported runtime

37+

boundary, regardless of the stored `/queue` setting.

38+39+

Use:

40+41+

- `/steer <message>` when you want to guide the active run right now.

42+

- `/queue steer` when you want future normal messages to steer active runs by

43+

default.

44+

- `/queue collect` or `/queue followup` when new messages should wait for a

45+

later turn instead of steering the active run.

46+47+

For queue modes and fallback behavior, see [Command queue](/concepts/queue) and

48+

[Steering queue](/concepts/queue-steering).

49+50+

## Sub-agents

51+52+

Use `/subagents steer` when the target is a child run:

53+54+

```text

55+

/subagents steer 2 focus only on the API surface

56+

```

57+58+

Top-level `/steer` does not select a sub-agent by id or list index. It always

59+

targets the current session's active run. See [Sub-agents](/tools/subagents) for

60+

sub-agent ids, labels, and control commands.

61+62+

## ACP sessions

63+64+

Use `/acp steer` when the target is an ACP harness session:

65+66+

```text

67+

/acp steer --session agent:main:acp:codex tighten the repro

68+

```

69+70+

See [ACP agents](/tools/acp-agents) for ACP session selection and runtime

71+

behavior.

72+73+

## Related

74+75+

- [Slash commands](/tools/slash-commands)

76+

- [Command queue](/concepts/queue)

77+

- [Steering queue](/concepts/queue-steering)

78+

- [Sub-agents](/tools/subagents)