Claude cli agent

Skill ImL1s/agent-cli-skills/skills/claude-cli-agent

Run Anthropic Claude Code headlessly via claude-exec.sh (claude -p). Use when the user asks for Claude Code, claude CLI, ultracode, Claude workflows, agent teams, or Claude subagents. -r is plan mode (clears skip-permissions); -T/--ultracode enables effort ultracode + multi-agent preamble; -P sets permission-mode. Orchestrator owns git (--no-git).From its SKILL.md

Install

npx -y skills add ImL1s/agent-cli-skills --skill claude-cli-agent

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.9 KB, 721 tokens by cl100k_base, as published. Nobody here has run it

Claude Code CLI (`claude -p`) Executor

When to use

User wants Claude Code / claude as the coding agent
Need ultracode, workflows, --agents, or worktree isolation from an orchestrator
Read-only second opinions (-r plan mode)

For cross-CLI councils, pair with multi-cli-spawn.

Wrapper

./claude-exec.sh -C /path/to/repo "Implement X; stop when done."
./claude-exec.sh -r -C /path/to/repo "Second opinion on this design"
./claude-exec.sh -T -C /path/to/repo "Large audit — use ultracode workflows"
./claude-exec.sh -e high -m sonnet -o json -f /tmp/bundle.txt
./claude-exec.sh -w task-branch --agents '{"reviewer":{"description":"Reviews","prompt":"You review code"}}' "…"

Flags: -m · -t · -C · -o text|json|stream-json · -f · -r (plan mode) · -T/--ultracode → --effort ultracode + multi-agent preamble · -P permission-mode · -e effort · -w worktree · --agents JSON · --allowed-tools · --disallowed-tools · --add-dir · --system-prompt · --append-system-prompt · --mcp-config · --strict-mcp-config · --bare · --ask-permissions · --no-git · -- passthrough.

Default executor uses --dangerously-skip-permissions. Setting -P or -r clears that skip. Prefer capturing stdout for reports. -C is absolutized before use.

Native multi-agent / ultracode

Mechanism	What it is	How to enable
Dynamic workflows	JS orchestration script fans out many subagents	Ask for a workflow, or keyword `ultracode`
ultracode	Session policy: xhigh effort + auto-workflow for substantive tasks	`-T` / `--effort ultracode` / interactive `/effort ultracode`
`--agents` JSON	Custom agent definitions for the session	`--agents '{...}'`
`claude agents`	Manage background agents	Separate subcommand
Agent teams	Small peer set that renegotiates live	Product feature; see Claude Code docs
`--worktree`	Isolated git worktree for the session	`-w [name]`

After a heavy ultracode session, drop back to normal effort for routine edits (token cost).

Pitfalls

--bare can break OAuth login for normal reviews — avoid unless you intend API-key-only
Do not put multi-KB $(cat brief) immediately after --allowedTools (parser confusion)
Plan mode cannot Write — if you need an answer file, allow Write or redirect stdout

Discipline

Exit code untrusted. Orchestrator owns git. This skill is not the OMX ask-claude artifact helper.

What ships with it: 4 files

17.5 KB alongside SKILL.md, 4 of them executable

lib/

common.shruns6.8 KB
parse_stream_json.pyruns3.5 KB
pty_run.pyruns1.7 KB

claude-exec.shruns5.4 KB

Gives 0 of the 12 instructions most agent orchestration skills give in 721 tokens

Counted across 848 of the 1,300 authors here whose files we hold, read 2026-09-06

Dispatch one agent per independent problem domainin 56 of 848, across 42 files
Run full test suite after integrationin 55 of 848, across 42 files
Verify fixes do not conflictin 40 of 848, across 32 files
Review each summary when agents returnin 40 of 848, across 31 files
Write a handoff document summarising the current conversationin 30 of 848, across 25 files
Reference existing artifacts by path or URLin 26 of 848, across 24 files
Give each agent a specific scopein 19 of 848, across 10 files
Give each agent a clear goalin 19 of 848, across 10 files
Include a suggested skills section in the documentin 18 of 848, across 16 files
Tailor the doc to the user argumentsin 18 of 848, across 15 files
Issue all subagent dispatches in the same responsein 17 of 848, across 11 files
Use git worktrees for isolationin 17 of 848, across 8 files

Said here and by no other author read

Run Anthropic Claude Code headlessly
Capture stdout for reports
Drop back to normal effort after sessions

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.