Skip to content
Agentmart
Submit a tool

runagent

Use when running applications, servers, or other long running commands in the background during debugging, testing, or development.

Skill by tsaarni

IndexedApache-2.0
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/tsaarni-runagent

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when running applications, servers, or other long running commands in the background during debugging, testing, or development.

Use cases: testing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/runagent

What's inside

Skill name
runagent
Description
Use when running applications, servers, or other long running commands in the background during debugging, testing, or development.
License
Apache-2.0
Instructions
86 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

`runagent` is installed and available in `PATH`. If not it can be run via `go run github.com/tsaarni/runagent/cmd/runagent@latest`

```
Usage: runagent <command> [flags]

Flags:
  -h, --help    Show context-sensitive help.
      --json    Output in JSON format.

Commands:
  run <command>
    Spawn a background process.
    -n, --name=NAME            Process name.
    -e, --env=KEY=VALUE,...    Environment variable KEY=VALUE (repeatable).
        --cwd=PATH             Working directory.

  ps [target]
    List processes or show process details.
    -a, --all                  Show details for all processes.

  logs <target>
    Read process log output.
        --type=TYPES           Event types to show (comma-separated: start,log,stats,stop).
        --time-range=FROM..TO  Time range (FROM..TO). FROM/TO: -5m (ago), 21:00:00, 2026-06-24T21:00:00, +2m (relative to FROM).
        --limit=N              Max events to return (first N from window).
        --last=N               Return last N matching events.
        --time-format=FORMAT   Timestamp format: time, datetime, none, or Go layout string.
    -f, --follow               Follow log output.
        --stream=STREAM        Filter by stream (stdout or stderr).

  kill <target>
    Send signal to process (default: SIGTERM).
        --signal=SIGNAL        Signal to send.

  delete [target]
    Kill process (if running) and delete its logs.
        --all                  Delete all processes.
        --stopped              Delete all processes that are not running (exited or killed).

  wait <target>
    Block until process exits.
        --timeout=DURATION     Timeout duration (e.g., 30s).

  daemon status
    Show daemon status.

  daemon start
    Start the daemon.

  daemon stop
    Stop the daemon.

  daemon clean
    Remove all runtime and state files.

Examples:
  runagent run -n web -- python3 -m http.server 8080
  runagent ps
  runagent logs web --last 20
  runagent wait web
  runagent delete web
```

Process States

- **Running** — process is alive
- **Exited** — process called exit() (check exit code for success/failure)
- **Killed** — terminated by signal

`runagent ps` shows process state, PID, command, uptime, CPU usage, memory usage, open file descriptors, disk I/O, etc.

`runagent logs` log line markers: │ = runagent control messages, ~ = resource stats, blank = process stdout/stderr. Stderr lines are shown in red when color is enabled. Timestamp format is customizable with --time-format.


## Rules

* Always name processes using `-n <name>`.
* For background jobs that complete on their own (such as long compilations), run: `runagent run -n <name> -- <compile-commands...>; runagent wait <name>; runagent logs <name> --last <num-lines>`.
* To wait for a process to finish, run `runagent wait <name>`. This returns immediately when the process exits, so you do not need to poll.
* Do not filter output when starting a process (avoid commands like `| grep ERROR`). Filter output later using `runagent logs`.
* `runagent run` automatically replaces dead processes that have the same name. If a process is still active, the command fails; stop or remove it first with `runagent kill <name>` or `runagent delete <name>`.
* When re-running recurring batch jobs prefer using same name in consequent runs, to avoid cluttering the `runagent ps` list with too many entries.
* After launching a process, run `runagent logs <name> --last 10` to confirm that it started properly.
* Keep processes alive after a test run unless instructed to remove them. They may be needed for debugging or follow-up testing.

From tsaarni/agent-skills, Apache-2.0.

More from tsaarni/agent-skills

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 23, 2026 (5 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from tsaarni

Cite this page

runagent skill. Agent Mart. https://agentmart.dev/skills/tsaarni-runagent (last updated 2026-10-24)