mode
Skill4 starsChanged 16 days agoMITView on GitHub
Get or set daemon operating mode - switch between default and unattended (blocks Stop events) - and turn this session's focus mode or manual-goals mode on or off
11 other projects use a version of this.
Adopted by
Most recently taken up by
- agoopintsai/shiprighttheir own version, first seen Nov 1, 2026
- eprouveze/claude-skillstheir own version, first seen Oct 20, 2026
- Scylla23/modelroutertheir own version, first seen Oct 20, 2026
- ProjectAJ14/eklavyatheir own version, first seen Oct 20, 2026
- DATA-AI-XYZ/Tandemtheir own version, first seen Oct 15, 2026
- lukasrepublic/agentic-foundrytheir own version, first seen Oct 11, 2026
Used with
Projects with it also use setup.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
VercelSkillsetup
Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".
Also in 3 projects that use this
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 11
- ProjectAJ14/eklavya★ 29same name
- YehudaFrankel/clankbrain★ 16same name
- LarryStanley/pi-delegate★ 4same name
- agoopintsai/shipright★ 2same name
- ahamedzoha/atelier★ 2same name
- ibahgat/oh-my-iflow★ 2same name
- lukasrepublic/agentic-foundry★ 1same name
- Scylla23/modelrouter★ 1same name
- StakeholderSensei/polygraph★ 1same name
- DATA-AI-XYZ/Tandem★ 0same name
1 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- .claude/agents/CLAUDE.mdCLAUDE.md
- CLAUDE/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- .claude/skills/CLAUDE.mdCLAUDE.md
- docs/CLAUDE.mdCLAUDE.md
- src/CLAUDE.mdCLAUDE.md
- tests/CLAUDE.mdCLAUDE.md
- acceptance-testSkill
- configureSkill
- docs-optimiseSkill
- docs-qaSkill
- hooks-daemonSkill
- issue-sdlcSkill
- releaseSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Follow its output exactly.
- Distinctive: line 71
When enabled, a mode interceptor runs before the handler chain on every event:
- Distinctive: line 74
SubagentStop events: NOT blocked (subagents should stop normally)
- Distinctive: line 81
When switching to unattended mode, you can include a custom message that gets appended to the block reason.
“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/mode - Daemon Mode Management
Run this first
Claude Code loads this markdown and nothing else — the file below is not run for you, and it holds the procedure to follow (Plan 00324):
bash "${CLAUDE_SKILL_DIR:-.claude/skills/mode}/invoke.sh" <the arguments the user typed>Follow its output exactly; what is written here is orientation.
Description
View or change the daemon's operating mode. Supports default (normal operation) and unattended (blocks Stop events to keep Claude working autonomously).
Usage
# Show current mode
/mode
/mode get
# Switch to unattended mode (blocks Stop events)
/mode unattended
# Switch to unattended with custom task message
/mode unattended finish the release and run all tests
# Switch back to default mode
/mode default
# Focus THIS session on one task (session mode; see focus_mode_trigger)
/mode focus <task>
# End focus (aliases: done, end, disable)
/mode focus off
# Manual goals for THIS session: no automated goal write or retirement clear (alias: off)
/mode goals manual
# Back to automated goals (alias: on)
/mode goals auto/mode goals manual|auto runs hooks-daemon goals off|on, the same as typing
hooks-daemon goals manual as a whole prompt; see CLAUDE/SessionModes.md (manual-goals).
/mode focus is the same as typing hooks-daemon focus <task> at the start of a prompt: it
changes only the current session, not the daemon-wide mode. It runs
hooks-daemon session-mode focus <task> / session-mode focus off; /mode get also prints
session-mode get. hooks-daemon mode-history lists every session's live modes and the recent
mode transitions, each with what triggered it.
Modes
| Mode | Behaviour |
default | Normal operation - all handlers process events as configured |
unattended | Blocks Stop events unconditionally to keep Claude working without interruption |
What Unattended Mode Does
When enabled, a mode interceptor runs before the handler chain on every event:
- Stop events: Blocked with a DENY response instructing Claude to continue working
- SubagentStop events: NOT blocked (subagents should stop normally)
- All other events: Pass through to normal handler chain unchanged
The interceptor includes re-entry protection (checks stop_hook_active flag) to prevent infinite loops.
Custom Messages
When switching to unattended mode, you can include a custom message that gets appended to the block reason. This lets you give Claude specific instructions about what to work on:
/mode unattended complete all remaining tasks in the planThe message appears in the Stop event denial reason, so Claude sees it when it tries to stop.
Python Path Detection
There is no interpreter to detect. The deployed wrapper resolves the fingerprint-keyed venv itself, and lives at the same place relative to the daemon root in both install modes:
# self-install (daemon development)
./bin/hooks-daemon <command>
# normal install
.claude/hooks-daemon/bin/hooks-daemon <command>IPC Protocol
The skill communicates with the daemon via Unix socket system requests:
Get mode:
{"event": "_system", "hook_input": {"action": "get_mode"}}Set mode:
{"event": "_system", "hook_input": {"action": "set_mode", "mode": "unattended", "custom_message": "optional"}}Session modes (one session):
{"event": "_system", "hook_input": {"action": "get_session_modes", "session_id": "<id>"}}
{"event": "_system", "hook_input": {"action": "enter_focus", "session_id": "<id>", "task": "<task>"}}
{"event": "_system", "hook_input": {"action": "exit_focus", "session_id": "<id>"}}Output
On get mode:
Mode: defaultOn set unattended:
Mode: unattended (changed)
Message: finish the releaseOn set default:
Mode: default (changed)Requirements
- Daemon must be running
- Python venv must be available (self-install or normal install)
Version
Introduced in: v2.17.0
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

