local-agent
VM0 local-agent CLI for delegating work to an already linked local Codex or Claude host. Use when an agent should list local-agent hosts or run tasks through `npx -p @vm0/cli zero local-agent`.
What this skill does
# VM0 Local Agent Local-agent lets this agent submit a natural-language job to a local-agent host that is already available in VM0. The host runs Codex or Claude Code in its configured environment, then returns the result to the current agent. Use it when the current sandbox is not the right execution environment, such as when the task needs a user's local repo, local files, local tools, or a specific workstation. ```bash npx -y -p @vm0/cli zero local-agent <command> ``` `-y` avoids npm's install prompt. Use `npx -p @vm0/cli ...` without `-y` only when an interactive prompt is acceptable. ## List Hosts List local-agent hosts only when the user asks to see hosts, when they ask to choose between hosts, or when an explicit host reference is ambiguous: ```bash npx -y -p @vm0/cli zero local-agent list ``` The output columns are: - `HOST ID`: host identifier shown for reference - `STATUS`: `online` hosts can accept jobs; `closed` hosts cannot - `BACKENDS`: supported agent backend, such as `codex` or `claude-code` - `LAST SEEN`: how recently the host checked in - `NAME`: the host name used by `run --host` Important: `run --host` takes the host `NAME`, not the host id. If multiple hosts have the same name, the run will fail as ambiguous. ## Run a Job When the user invokes `/local-agent <task>`: - If they explicitly name a host, run the task with `--host "<host-name>"`. - If they do not specify a host, do not list hosts first; run directly on any online host. - Treat everything after `/local-agent` as the remote job prompt, preserving the user's intent. Run on any online host: ```bash npx -y -p @vm0/cli zero local-agent run "Check the repo status and summarize the current branch." ``` Run on a specific host by name: ```bash npx -y -p @vm0/cli zero local-agent run \ --host "<host-name>" \ --timeout 3600 \ "In /path/to/repo, run the tests for package X and report failures with file paths." ``` Options: - `--host <name>`: send the job to a named host from `zero local-agent list` - `--timeout <seconds>`: maximum time to wait; defaults to `7200` The command queues the job, polls until it finishes or times out, prints the remote output, and exits non-zero if the remote job fails. ## Inspect Runs Use run inspection commands when the user asks for previous local-agent jobs, when `run` times out, or when you need to check a job without re-running it. List recent local-agent runs: ```bash npx -y -p @vm0/cli zero local-agent runs list ``` Useful filters: ```bash npx -y -p @vm0/cli zero local-agent runs list --status running npx -y -p @vm0/cli zero local-agent runs list --host "<host-name>" --limit 10 npx -y -p @vm0/cli zero local-agent runs list --json ``` The list output shows job id, status, host, backend, creation time, and a prompt preview. Use the job id for status and result lookups. Check one job's status: ```bash npx -y -p @vm0/cli zero local-agent runs status <job-id> ``` Print one job's final result: ```bash npx -y -p @vm0/cli zero local-agent runs result <job-id> ``` `runs result` prints successful job output to stdout. For failed jobs it prints the remote error and exits non-zero. For queued or running jobs, use `runs status` first or wait before asking for the result. For multiline prompts, pass one shell argument: ```bash prompt=$(cat <<'EOF' Task: inspect the current checkout and find why the build is failing. Expected output: - concise root cause - exact commands run - changed files, if any - remaining risks EOF ) npx -y -p @vm0/cli zero local-agent run --host "<host-name>" --timeout 7200 "$prompt" ``` ## Prompting Guidelines - Include the exact task, expected output, and any relevant repo paths or branches. - State whether the local agent may edit files, run tests, install dependencies, or only inspect. - Do not assume the remote host has the same files as the current sandbox. Include enough context for the host's configured working directory. - Keep prompts below 60,000 characters. - Ask for concise output. Remote job output keeps only the latest output up to the CLI limit, so request summaries plus file paths or commands rather than large logs. - Treat local-agent as execution on another machine. Do not send secrets or private data unless the task requires it and the user authorized it.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.