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 antoinedc/MantaUI --skill mantagit clone --depth 1 https://github.com/antoinedc/MantaUIWrote 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/antoinedc/mantaui/manta)<a href="https://agentmods.dev/skills/antoinedc/mantaui/manta"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta/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/antoinedc/mantaui/manta"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta.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 | $0.00069 | $0.03520 |
| Opus 5 | $0.00034 | $0.01760 |
| Sonnet 5 | $0.00014 | $0.00704 |
| Haiku 4.5 | $0.00007 | $0.00352 |
Grade D, and why
manta scanned grade D with 3 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 10d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
and ask them to run it or approve it — do not run sudo yourself unless the Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://mantaui.com/install.sh | bash Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `for c in curl tar tmux git; do command -v $c >/dev/null || echo "missing: $c"; done` How it starts
The opening of the file, as written. The whole thing — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an AI coding agent installing Manta UI's box server on behalf of your user. Follow this document exactly. Do not improvise install steps: the installer script is the single mechanism; your job is the interview, running it with the right knobs, verifying, and reporting.
Safety invariants (never violate)
- NEVER delete or regenerate
~/.manta/auth.json. It is the box identity; regenerating it unpairs every device the user has. - NEVER edit
~/.config/opencode/opencode.jsoncdestructively — the installer merges what it needs; leave it alone. - NEVER run the installer as root or with sudo. It is a per-user, user-space install (systemd --user + linger). It never installs system packages. If a prerequisite is missing (e.g. tmux), show the user the exact install command and ask them to run it or approve it — do not run sudo yourself unless the user explicitly approves that command.
- NEVER install Node.js or build tools — the installer ships its own Node runtime inside the release tarball.
- Re-running the installer is safe: it upgrades in place and preserves identity.
Step 1 — interview (ask BEFORE running anything)
Ask the user exactly this question:
- "Am I running on the Linux box you want to install Manta on, or should I
install on a remote box over SSH?" If remote: ask for
user@host(and key/port if needed), then run every command below throughssh user@host '…'. Verify SSH works withssh user@host 'echo ok'first.
Do NOT ask about chat mode (always installed), model providers, or projects — the desktop app's onboarding handles those.
macOS box (Apple Silicon) — read this if the answer was "this Mac"
The installer supports macOS as a box OS only on Apple Silicon (arm64)
machines running a recent macOS. Intel Macs are not supported as a box —
the installer will refuse with a clear message rather than silently failing;
if the user wants to USE Manta on an Intel Mac, point them at the desktop app
instead (https://mantaui.com/downloads/Manta-latest.dmg).
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.
- 10d ago First seen · 235 lines · 69 tokens per session scan D bf85f87203cc
manta is a skill published in the GitHub repository antoinedc/MantaUI (12 stars, last pushed 3d ago), licensed MIT. It adds 69 tokens to every session and 3,520 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it D with 3 findings (asks for root, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
im-adapter
Format responses for instant messaging platforms such as Lark, DingTalk, WeCom, Slack, and Telegram. Controls response length, Markdown formatting, tone, group chat behavior, and the [PASS] protocol. Also covers sending images and files via [SENDIMAGE]/[SENDFILE] markers on channels that support media (currently WeCom…
general
Handle everyday conversation, answer questions, manage files, take notes, run scripts, and maintain persistent memory across sessions. Use when the user asks a general question, requests file operations, wants to brainstorm ideas, needs to-do tracking, asks you to remember something, or requests skill search and…
message-push
Send proactive messages to IM groups or individual users via the gateway Send API. Use when the user says to send, post, notify, forward, or tell someone a message on Feishu, Slack, Telegram, Discord, DingTalk, or WeCom.
multi-bot
Coordinates responses between multiple GolemBot instances in a shared fleet. Use when the bot operates in a group chat with other bots, needs to decide whether to respond or pass, or must call a peer bot's API to fetch cross-domain data.
task-manager
Creates and manages scheduled tasks, cron jobs, recurring reminders, and timers via the Task HTTP API. Use when the user asks to schedule something, set a recurring reminder, run a periodic check, or manage existing scheduled tasks.
data-analysis
Loads CSV, Excel, and JSON data files, performs statistical analysis, and generates charts and reports. Use when the user asks to analyze a dataset, compute statistics, create visualizations, find trends, or produce a data report.