How do I install the Agent Tail skill?

Run npx skillmds@latest add gillkyle/agent-tail in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agent Tail safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agent Tail?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agent Tail free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agent Tail?

gillkyle (@gillkyle) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agent Tail by gillkyle — Agent Skill

Name: Agent Tail
Rating: 4.2 (5 reviews)
Author: gillkyle

File contents

agent-tail

Pipes browser console output and dev server stdout/stderr to plain log files on disk so you can grep, tail, and read them directly.

Reading logs

agent-tail tail -n 200
agent-tail tail browser -n 50
agent-tail tail -f
grep -ri "error\|warn" tmp/logs/latest/
tail -50 tmp/logs/latest/browser.log
tail -f tmp/logs/latest/*.log

Use plain tail if direct file paths are easier. Use agent-tail tail if you want the CLI to resolve the latest session for you and forward flags like -f and -n 200 directly to tail.

Log files

Logs live in tmp/logs/latest/ (a symlink to the newest session directory, stable across restarts):

File	Source
`browser.log`	Browser `console.*`, unhandled errors/rejections (Vite or Next.js plugin)
`<name>.log`	stdout/stderr of a named service (CLI)
`combined.log`	All services interleaved, prefixed with `[name]` (CLI)

Log format

[HH:MM:SS.mmm] [LEVEL  ] message (source-url)
    stack trace indented with 4 spaces

Levels are uppercased and padded to 7 characters (LOG , WARN , ERROR ).

CLI

agent-tail run 'fe: npm run dev' 'api: uv run server'   # run multiple services
agent-tail wrap api -- uv run fastapi-server              # add to existing session
agent-tail init                                           # create session only
tail -f tmp/logs/latest/*.log                             # tail directly by path
agent-tail tail -f                                        # tail all logs in latest session
agent-tail tail browser -n 50                             # tail one log in latest session

Flags (work with all commands):

--log-dir <dir>       Log directory (default: tmp/logs)
--max-sessions <n>    Max sessions to keep (default: 10)
--no-combined         Don't write combined.log
--exclude <pattern>   Exclude lines matching pattern (repeatable, /regex/ or substring)
--mute <name>         Mute service from terminal + combined.log (still logs to <name>.log)

See references/cli-reference.md for full details.

Setup

npm install -D agent-tail

CLI (any stack): Add agent-tail run 'name: command' to your dev script
Vite: See references/setup-vite.md
Next.js: See references/setup-nextjs.md (config wrapper + layout script + API route)

Agent instructions

Add to your project's CLAUDE.md, .cursorrules, or equivalent:

## Dev Logs

All dev server output is captured to `tmp/logs/`. The latest session
is symlinked at `tmp/logs/latest/`.

When debugging, check logs before guessing about runtime behavior:

    grep -ri "error\|warn" tmp/logs/latest/
    tail -50 tmp/logs/latest/browser.log
    agent-tail tail browser -n 50

.gitignore

Add tmp/ to .gitignore. agent-tail warns on startup if the log directory isn't gitignored.

---name: agent-taildescription: Capture browser console logs and dev server output to files with agent-tail. Use when debugging runtime errors, checking console output, tailing or diagnosing logs, or setting up Vite/Next.js log capture.---# agent-tailPipes browser console output and dev server stdout/stderr to plain log files on disk so you can `grep`, `tail`, and read them directly.## Reading logs```bashagent-tail tail -n 200agent-tail tail browser -n 50agent-tail tail -fgrep -ri "error\|warn" tmp/logs/latest/tail -50 tmp/logs/latest/browser.logtail -f tmp/logs/latest/*.log```Use plain `tail` if direct file paths are easier. Use `agent-tail tail` if you want the CLI to resolve the latest session for you and forward flags like `-f` and `-n 200` directly to `tail`.## Log filesLogs live in `tmp/logs/latest/` (a symlink to the newest session directory, stable across restarts):| File | Source ||------|--------|| `browser.log` | Browser `console.*`, unhandled errors/rejections (Vite or Next.js plugin) || `<name>.log` | stdout/stderr of a named service (CLI) || `combined.log` | All services interleaved, prefixed with `[name]` (CLI) |## Log format```[HH:MM:SS.mmm] [LEVEL ] message (source-url) stack trace indented with 4 spaces```Levels are uppercased and padded to 7 characters (`LOG `, `WARN `, `ERROR `).## CLI```bashagent-tail run 'fe: npm run dev' 'api: uv run server' # run multiple servicesagent-tail wrap api -- uv run fastapi-server # add to existing sessionagent-tail init # create session onlytail -f tmp/logs/latest/*.log # tail directly by pathagent-tail tail -f # tail all logs in latest sessionagent-tail tail browser -n 50 # tail one log in latest session```**Flags** (work with all commands):```--log-dir <dir> Log directory (default: tmp/logs)--max-sessions <n> Max sessions to keep (default: 10)--no-combined Don't write combined.log--exclude <pattern> Exclude lines matching pattern (repeatable, /regex/ or substring)--mute <name> Mute service from terminal + combined.log (still logs to <name>.log)```See `references/cli-reference.md` for full details.## Setup```bashnpm install -D agent-tail```- **CLI (any stack):** Add `agent-tail run 'name: command'` to your dev script- **Vite:** See `references/setup-vite.md`- **Next.js:** See `references/setup-nextjs.md` (config wrapper + layout script + API route)## Agent instructionsAdd to your project's `CLAUDE.md`, `.cursorrules`, or equivalent:```markdown## Dev LogsAll dev server output is captured to `tmp/logs/`. The latest sessionis symlinked at `tmp/logs/latest/`.When debugging, check logs before guessing about runtime behavior: grep -ri "error\|warn" tmp/logs/latest/ tail -50 tmp/logs/latest/browser.log agent-tail tail browser -n 50```## .gitignoreAdd `tmp/` to `.gitignore`. agent-tail warns on startup if the log directory isn't gitignored.

Frequently asked questions

Agent Tail