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 dongdong306/hailux --skill default_helpgit clone --depth 1 https://github.com/dongdong306/hailuxWrote 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/dongdong306/hailux/default_help)<a href="https://agentmods.dev/skills/dongdong306/hailux/default_help"><img src="https://agentmods.dev/badge/skills/dongdong306/hailux/default_help/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/dongdong306/hailux/default_help"><img src="https://agentmods.dev/badge/skills/dongdong306/hailux/default_help.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 | $0.00039 | $0.03835 |
| Opus 5.5 | $0.00016 | $0.01534 |
| Sonnet 5.5 | $0.00008 | $0.00767 |
| Haiku 4.5 | $0.00004 | $0.00383 |
Grade B, and why
help scanned grade B with 1 finding 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 13d 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.
"sudo *" = "allow" # stop asking for sudo commands How it starts
The opening of the file, as written. The whole thing — 404 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hailux User Guide
hailux is a terminal AI coding assistant that supports streaming conversations, tool calls, Skills, custom commands, MCP server integration, and subagent delegation.
Command Line Interface
| Command | Description |
|---|---|
hailux |
Launch the interactive TUI |
hailux --path <dir> |
Set the working directory |
hailux --rebuild-db |
Rebuild the database (clears all history), used when migration fails; the old database is backed up first |
hailux --yolo |
YOLO mode: skip all permission confirmations (works in both TUI and non-interactive mode) |
hailux run <message> |
Non-interactive mode: send a single message and print the reply (reads from stdin if the message is omitted) |
hailux run --model <provider/model> |
Override the model specified in config |
hailux run --no-tools |
Disable all tool calls (including MCP), pure chat mode |
Permissions
By default, operations inside the working directory are allowed without confirmation:
- bash commands,
read/edit/write/grep/globinside the working directory run freely - reading sensitive files such as
.env/.env.localasks for confirmation (.env.exampleis allowed) - any operation touching content outside the working directory (e.g.
cat ../secrets.txt,rm /tmp/x, editing a file outside the project) asks for confirmation — including bash sub-commands chained with&&/;/|, redirection targets, andcdtargets outside the working directory - bash commands that cannot be statically analyzed (sudo, sh/bash, python/node interpreters, eval, xargs, ...) ask for confirmation on first use per session — built-in defaults never silently authorize them
- MCP tools always ask for confirmation
Note: The permission layer is convenience gating, not a security sandbox. bash is arbitrary code execution by nature; the static analysis of shell commands is best-effort (e.g. unknown
$variablescannot be resolved). Treat--yoloand permission rules as trust decisions.
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.
- 13d ago Changed · +4 lines scan A → B 74052ce52edd
- 1mo ago First seen · 400 lines · 39 tokens per session scan A 2766e8271798
help is a skill published in the GitHub repository dongdong306/hailux (5 stars, last pushed 3d ago), licensed MIT. It adds 39 tokens to every session and 3,835 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-08.
Other skills, from other repositories
ast-grep
Use for ast-grep project setup, rule authoring, rule debugging, and structural search workflows.
skill-installer
Install VT Code skills into /.agents/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo, including private repos.
cmd-review
Review the current diff or selected files (usage: /review [instructions | --last-diff | --target | --file | files...] [--style ]).
vtcode-companion
Delegate repository investigation, coding plans, explicitly requested implementation, and verification review to a connected local VT Code Companion MCP service.
unifly
Inspect and manage UniFi Network infrastructure with the unifly CLI: devices, clients, VLANs, WiFi, firewall, VPN, switch ports, site settings, and cloud fleets. Use for UniFi network configuration, monitoring, and troubleshooting, not UniFi Protect camera or recording management.
unifly
Inspect and manage UniFi Network infrastructure with the unifly CLI: devices, clients, VLANs, WiFi, firewall, VPN, switch ports, site settings, and cloud fleets. Use for UniFi network configuration, monitoring, and troubleshooting, not UniFi Protect camera or recording management.