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/pyworkload/3x-ui-mcp/smart-routingnpx skills add pyworkload/3x-ui-mcp --skill smart-routinggit 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/smart-routing)<a href="https://agentmods.dev/skills/pyworkload/3x-ui-mcp/smart-routing"><img src="https://agentmods.dev/badge/skills/pyworkload/3x-ui-mcp/smart-routing.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 | $0.00090 | $0.01326 |
| Opus 5 | $0.00045 | $0.00663 |
| Sonnet 5 | $0.00018 | $0.00265 |
| Haiku 4.5 | $0.00009 | $0.00133 |
Grade A, and why
smart-routing 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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Routing
Routing rules live in the saved Xray template, and every rule tool
(add_routing_rule, update_routing_rule, remove_routing_rule) rewrites that
whole template read-modify-write. Two consequences: back up before a session of
edits, and do not interleave rule edits with update_xray_template, which replaces
the document outright and will discard rules added in between.
Before anything else
Sniffing must be enabled on the inbound. Without it the router sees only IP
addresses and every domain rule misses — silently, with no error anywhere. The
default sniffing this server sends on create_inbound has it on; an inbound
created elsewhere may not. Check get_inbound first when domain rules "do nothing".
Rule order is the whole game
get_routing_rules
Rules are evaluated top to bottom and the first match wins. The classic failure is
a broad rule above a narrow one — geosite:category-ads above a specific
domain:ads.example.com exception means the exception never runs.
add_routing_rule appends by default; pass index to place a rule where it will
actually be reached.
get_routing_rules also returns the balancer definitions, because a rule's
balancerTag is meaningless without them.
Validate geo tokens before saving
geosite: and geoip: tokens that do not exist do not error — they just never
match:
validate_geodata_tokens tokens="geosite:category-ads-all,geoip:private" kind="domain"
list_geodata_files, list_geodata_categories and list_geodata_entries browse
what the server's databases actually contain, which is the reliable way to find the
right category name rather than guessing from a blog post. update_geofile
refreshes the databases; stale geo data is a common cause of rules that used to
work.
Common shapes
Direct for local traffic, so it never leaves through the proxy:
add_routing_rule rule={"type":"field","outboundTag":"direct","ip":["geoip:private"]} index=0
Block a category:
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 · 134 lines · 90 tokens per session scan A 4c381c89b8fd
smart-routing is a skill published in the GitHub repository pyworkload/3x-ui-mcp (5 stars, last pushed 4d ago), licensed MIT. It adds 90 tokens to every session and 1,326 once invoked, about $0.0005 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
opik-diagnose
Surface the Opik traces worth a developer's attention, ranked by signal — errors, failed tool calls, latency, regressions, and low online-eval scores — plus Diagnostics issues. Reads live/production traces via the SDK (searchtraces and agentinsights) and works with no MCP; uses the MCP issue entity when connected.…
cortex-explore-memory
Explore the memory system's state, find gaps in knowledge, assess coverage, and get diagnostic information. Use when the user asks 'what does my memory look like', 'show me memory stats', 'what am I missing', 'how good is my knowledge', 'memory health', 'show coverage', 'find gaps', 'what topics are weak', or when you…
qmclaw-workbench
使用 OpenQuantum 的 QMClaw Local Tool 对超导量子比特 S21、能谱、Rabi、Ramsey、T1、SingleShot、DRAG、π 脉冲、功率偏移、Delta 和 RB 等 13 类测控实验做有界、确定性的本地模拟,并组织单比特调校工作流。用于实验规划、接口联调、教学和无硬件预检;不连接 LabRAD/lqms、真实仪器或量子云,不修改校准参数,也不替代 Scientific Validator。.
quick-capture
Use this skill to drop a new task into GSD Task Manager from any AI assistant that has the gsd-mcp-server connected.
xmemo-vscode
Use XMemo from VS Code (and forks like Cursor/Windsurf) to recall project context, capture engineering decisions, preserve handoff state, and connect to the hosted XMemo MCP endpoint safely.
event-staffing-ordering
Order W-2 event staff for US/CA events through TempGuru.