Atomic Agent is a local-first AI agent that runs its control loop and state on a user's machine while using local or cloud models. It drives browsers, edits files, runs approved commands, remembers context, schedules follow-ups, and connects to external tools, with the catalogue providing skills for its use.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/atomicbot-ai/atomic-agent/gog-workspacenpx skills add AtomicBot-ai/atomic-agent --skill gog-workspacegit clone --depth 1 https://github.com/AtomicBot-ai/atomic-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/atomicbot-ai/atomic-agent/gog-workspace)<a href="https://agentmods.dev/skills/atomicbot-ai/atomic-agent/gog-workspace"><img src="https://agentmods.dev/badge/skills/atomicbot-ai/atomic-agent/gog-workspace.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00068 | $0.01732 |
| Opus 5 | $0.00034 | $0.00866 |
| Sonnet 5 | $0.00014 | $0.00346 |
| Haiku 4.5 | $0.00007 | $0.00173 |
Grade A, and why
gog-workspace scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gog workspace
Use gog as the Google Workspace gateway. Prefer gog over raw Google APIs.
Default Flow
For normal Gmail/Drive/Calendar reads, do not run setup/status/auth diagnostics first.
- Run the requested read command directly with safe
gog --no-inputand--enable-commandsflags. - If that read succeeds, answer the user from the result.
- If that read fails with installation, auth, account, credentials, or keyring errors, then run the diagnostic check below or tell the user the exact auth action from the command error.
Never call skill.run_script for check-gog.sh as a preflight step just because this is the first Workspace task in a session.
For unread Gmail, use this exact command shape. Do not invent flags:
[{ "tool": "os.shell.run", "args": { "cmd": "gog", "args": ["--json", "--no-input", "--account", "<email>", "--wrap-untrusted", "--gmail-no-send", "--enable-commands", "gmail.search,gmail.get", "gmail", "search", "is:unread newer_than:7d", "--max", "10"] } }]
Gmail search syntax is a positional <query> argument. Do not use --unread; that flag does not exist. Prefer the canonical verb gmail search over the aliases gmail list, gmail ls, gmail find, or gmail query unless the user explicitly asks for an alias.
Calendar event listing must use the canonical command calendar events with --enable-commands calendar.events. Do not use calendar list, calendar ls, or --enable-commands calendar.list; gog resolves those aliases to calendar events internally and rejects the call unless calendar.events is enabled.
Diagnostics
Run the non-interactive diagnostic check only when the user asks to check setup/status, after a real gog read command reports installation/auth/account/credentials/keyring failure, or after the user says they changed setup/auth:
[{ "tool": "skill.run_script", "args": { "skill": "gog-workspace", "script": "check-gog.sh" } }]
Outcome map:
status: healthy-> proceed withgog.status: install_required-> stop and tell the user how to installgog.status: auth_required-> stop and tell the exact auth commands from the check output. Prefergog auth credentials <credentials.json>followed bygog auth add <email>when those are shown; use the user's known email if available.status: api_required-> tell the user which Google API must be enabled, then stop.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 144 lines · 68 tokens per session scan A 15816d9ab37c
gog-workspace is a skill published in the GitHub repository AtomicBot-ai/atomic-agent (2,466 stars, last pushed yesterday), licensed MIT. It adds 68 tokens to every session and 1,732 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
excel-spreadsheets
Create, modify, and manage Excel spreadsheets.
google-workspace-ops
Operate across Google Workspace tools with reusable patterns for docs, sheets, calendar, drive, and gmail.
xlsx
Create, inspect, edit, restructure, convert, and validate XLSX workbooks and CSV files using Qubicl's pinned local spreadsheet tools. Use for formulas, tables, charts, formatting, workbook structure, and CSV interoperability.
google-workspace-ops
Google Workspace CLI automation via gogcli. Gmail, Calendar, Drive, Docs, Slides, Sheets, and 9 more services. JSON-first output, composable pipelines.
lark-cli
Lark/Feishu (飞书) CLI tool. Use when the user asks to do anything on Lark/Feishu — send or read chat messages (IM), read or edit docs/sheets/slides/wiki, work with multi-dimensional tables (Base/bitable), manage calendar/meetings/video-conference, mail, contacts, tasks, approvals, OKR, drive files, minutes…
data-report
Turns CSV, Excel, or JSON data into a polished visual report page.