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 skills add adbertram/cli-tools --skill roomba-cligit clone --depth 1 https://github.com/adbertram/cli-toolsWrote 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/adbertram/cli-tools/roomba-cli)<a href="https://agentmods.dev/skills/adbertram/cli-tools/roomba-cli"><img src="https://agentmods.dev/badge/skills/adbertram/cli-tools/roomba-cli/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/adbertram/cli-tools/roomba-cli"><img src="https://agentmods.dev/badge/skills/adbertram/cli-tools/roomba-cli.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00115 | $0.01681 |
| Opus 5.5 | $0.00046 | $0.00672 |
| Sonnet 5.5 | $0.00023 | $0.00336 |
| Haiku 4.5 | $0.00012 | $0.00168 |
Grade A, and why
roomba-cli 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 today.
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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<quick_start>
The roomba CLI follows this pattern:
roomba <command-group> <action> [arguments] [options]
| Task | Command |
|---|---|
| Quick status check | roomba status |
| List robots | roomba robots list --table |
| Start cleaning | roomba control start |
| Stop cleaning | roomba control stop |
| Send to dock | roomba control dock |
| Find robot (beep) | roomba control find |
| Evacuate bin | roomba control evac |
| Detailed status | roomba control status |
| </quick_start> |
<essential_principles>
MANDATORY: Consult the adjacent usage.json at <cli-tools-root>/_repo/skills/<tool>-cli/usage.json before executing ANY roomba command.
This file contains complete command syntax, all arguments, all options, and usage instructions for every command. Never guess at command syntax.
<reference_index>
usage.json -- Complete command tree with arguments, options, defaults, and usage instructions for every command.
</reference_index>
<success_criteria>
- Command executes without error
- Output is displayed in requested format
- Correct command and flags used (verified against usage.json) </success_criteria>
<cloud_reads>
Verified 2026-10-20 against successful authorized app traffic and live CLI reads.
cloud import-capture PRIVATE_PATH imports temporary AWS credentials through the
CLI-tools secret manager, cloud inventory, and exact supported successful GET
request templates. The decoder runs through pinned mitmproxy 12.2.3 via uv.
It requires a successful app /v2/login in the capture. A matching Gigya session
is verified before enabling automatic AWS/MQTT renewal. Captured old login proofs
and account/robot passwords are never retained. Requests must match that
login's access key, or another captured login with identical Cognito identity and
robot/deployment membership. Keep captures and raw output in private non-synced storage.
Use cloud robots list to inspect multiple captured vacuums, cloud select NAME_OR_ID
for the local cloud default, or cloud read history --robot NAME_OR_ID for explicit
selection. Robot history/setup templates are isolated by exact cloud robot ID.
cloud read schedule and cloud read entitlements are shared household/account
reads. A resource absent from that vacuum's capture fails explicitly. Local LAN
configuration and robot passwords are preserved. Temporary tokens renew automatically
when a verified reusable Gigya session was imported. Rejection or expiration of
that underlying session requires a fresh authorized app login capture.
Label this evidence Local source. Cloud data never establishes live device status or robot execution. The existing LAN path remains separate. This group supports both captured vacuum models; model-specific repair guidance still requires the appropriate hardware source. See the CLI README for redacted provenance and operating limits. Consult the adjacent generated usage map before commands. </cloud_reads>
<read_surface>
cloud status --robot NAME_OR_ID reports model-specific observed telemetry with
per-field metadata age; snapshots never prove current physical online status.
cloud schedules --robot NAME_OR_ID reads both household and stored robot
schedules. Household membership must match the selected robot. Empty household
schedules do not establish that stored robot schedules are empty.
cloud favorites list/get reads actual favorite definitions and enforces exact
robot membership. Favorites are never started by a read command.
What ships with it
1 file 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.
- today First seen · 149 lines · 115 tokens per session scan A 139ec39f2194
roomba-cli is a skill published in the GitHub repository adbertram/cli-tools (6 stars, last pushed today), licensed MIT. It adds 115 tokens to every session and 1,681 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-11-07.
roomba-cli compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| roomba-cliadbertram/cli-tools | Skill | 6 | today | MIT | A | 115 | Claude Code, Codex |
| opencli-autofixjackwener/OpenCLI | Skill | 31k | yesterday | Apache-2.0 | A | 61 | Claude Code |
| surfnicobailon/surf-cli | Skill | 684 | yesterday | MIT | A | 53 | Claude Code, Codex |
| rdc-cliBANANASJIM/rdc-cli | Skill | 193 | 14d ago | MIT | A | 82 | Claude Code, Codex |
| pixiv-cli-ciFlanChanXwO/pixiv-cli | Skill | 132 | yesterday | MIT | A | 61 | Codex |
| acrawl-mcpMingye-Lu/AgenticCrawler | Skill | 15 | yesterday | MIT | A | 200 | Claude Code, Codex |
| debug-hubrexleimo/aios | Skill | 62 | yesterday | MIT | B | 66 | Claude Code, Codex |