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 pyworkload/3x-ui-mcp --skill diagnose-connectivitygit clone --depth 1 https://github.com/pyworkload/3x-ui-mcpWrote 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/pyworkload/3x-ui-mcp/diagnose-connectivity)<a href="https://agentmods.dev/skills/pyworkload/3x-ui-mcp/diagnose-connectivity"><img src="https://agentmods.dev/badge/skills/pyworkload/3x-ui-mcp/diagnose-connectivity.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 5.1 | $0.00083 | $0.01650 |
| Opus 5 | $0.00042 | $0.00825 |
| Sonnet 5 | $0.00017 | $0.00330 |
| Haiku 4.5 | $0.00008 | $0.00165 |
Grade A, and why
diagnose-connectivity 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 2d 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Diagnosing a connection failure
Work from the cheapest, most common cause outward. Most reports resolve in the first two steps, and every step below is a read-only call.
Step 0 — scope it
Ask, or determine from the panel, which of these it is:
- Everyone, everywhere → the core or the box (step 1)
- Everyone on one inbound → the inbound (step 2)
- One client → that client's limits (step 3)
- Everyone, but only for one destination → routing or an outbound (step 4)
get_online_clients answers this immediately: a normal-looking list means the
core is up and serving, so skip straight to step 3.
Step 1 — the core and the box
server_status
get_xray_logs count=100
The single most common panel-wide outage is a template that Xray refuses to load:
the panel saves it happily, the core then fails to start, and every client drops.
get_xray_logs names the offending field. Confirm by comparing what is running
with what is saved — get_xray_config reports the running core, get_xray_template
the saved document; if the config call fails while the template reads fine, the
core is down.
Fix the template and restart_xray. If the panel itself is unreachable, nothing
here will answer at all — that is a host or firewall problem, not a config one.
Step 2 — the inbound
get_inbound id=<id>
Check, in this order: enable is true; port is what clients are actually
dialling and nothing else took it; expiryTime (Unix milliseconds, 0 = never)
has not passed; total (bytes, inbound-wide quota) is not exhausted. A disabled
or expired inbound looks exactly like a network failure from the client side.
For a REALITY inbound whose settings are unchanged but which stopped getting
through, the camouflage target is the suspect — see the reality-inbound skill.
Step 3 — the client
get_client email="alice"
get_client_traffic email="alice"
get_last_online
Five things disable a client that is otherwise correctly configured:
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.
- 2d ago First seen · 153 lines · 83 tokens per session scan A 109f33716d6a
diagnose-connectivity is a skill published in the GitHub repository pyworkload/3x-ui-mcp (5 stars, last pushed 5d ago), licensed MIT. It adds 83 tokens to every session and 1,650 once invoked, about $0.0004 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 2026-09-05.
Other skills, from other repositories
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.