new-claude-at
Skill1 starsChanged 7 months agoMITView on GitHub
Open a new Konsole window running a fresh Claude Code CLI session at an arbitrary path. Use when the user asks for "a new claude at <path>", "claude in <other repo>", "spawn claude over in <dir>", or similar. Unlike `new-claude-here`, this takes an explicit target directory. Supports optional seed prompt and self-destruct.
In this repository13 other files it gives its agents.
- claude-and-shellSkill
- claude-gridSkill
- clipboard-handoverSkill
- handoverSkill
- handover-with-tasksSkill
- new-claude-hereSkill
- new-workspaceSkill
- sideclaudeSkill
- spawn-planning-repoSkill
- start-from-handoverSkill
- start-leftoverSkill
- terminal-hereSkill
- vscode-hereSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 29
If a seed prompt is provided:
- Very specific: line 43
If the caller requested self-destruct, follow the same pattern as new-claude-here:
- Distinctive: line 46
Verify a new Konsole PID appeared after launch.
- Distinctive: line 47
If verified, warn the user and close the current Konsole:
- Distinctive: line 51
If no new PID appeared, ABORT — do not close anything.
- Distinctive: line 58
If self-destruct is pending
- Distinctive: line 64
Do not block waiting on the Konsole process.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
New Claude Session At Path
Spawn a sibling Claude Code session in a new Konsole window at an arbitrary directory the user specifies. Companion to new-claude-here (which uses $PWD) — use this one when the target is somewhere else on disk.
Inputs
- Target path — absolute path, or a path relative to the current repo. Required. If the user hasn't given one, ask. Expand
~and resolve to an absolute path before launching. - Seed prompt (optional) — an initial prompt or slash command to run in the new session (e.g.
/brainstorm-solutions:deep-research). Passed as a positional argument toclaude. - Self-destruct (optional) — if the caller says "close this one", "self-destruct", or "hand off", close the current Konsole after the new one is verified. Default: keep current open.
Steps
1. Resolve and verify target
target=$(readlink -f "<given-path>")
[ -d "$target" ] || { echo "not a directory: $target"; exit 1; }2. Launch new session
IMPORTANT: When passing a seed prompt, wrap claude in bash -c so that Konsole sets the working directory correctly before Claude starts. Without this wrapper, Claude may open in the wrong directory.
If a seed prompt is provided:
setsid konsole --workdir "$target" -e bash -c 'claude "<seed-prompt>"' >/dev/null 2>&1 &Otherwise:
setsid konsole --workdir "$target" -e claude >/dev/null 2>&1 &setsid + background detaches so the new window survives the current session exiting.
3. Self-destruct (if requested)
If the caller requested self-destruct, follow the same pattern as new-claude-here:
- Snapshot existing Konsole PIDs:
before=$(pgrep -x konsole). - Verify a new Konsole PID appeared after launch.
- If verified, warn the user and close the current Konsole:
(sleep 3 && kill -TERM $(ps -o ppid= -p $PPID | tr -d ' ')) & - If no new PID appeared, ABORT — do not close anything.
4. Report
Tell the user:
- The absolute path the new session was launched in
- Whether a seed prompt was passed
- If self-destruct is pending
Notes
- The new session is standalone — no shared memory with this one.
- For full context transfer without Junction, run
session-transfer:write-handoverfirst. - Do not block waiting on the Konsole process.
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.

