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/reality-inboundnpx skills add pyworkload/3x-ui-mcp --skill reality-inboundgit 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/reality-inbound)<a href="https://agentmods.dev/skills/pyworkload/3x-ui-mcp/reality-inbound"><img src="https://agentmods.dev/badge/skills/pyworkload/3x-ui-mcp/reality-inbound.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 | $0.00094 | $0.01512 |
| Opus 5 | $0.00047 | $0.00756 |
| Sonnet 5 | $0.00019 | $0.00302 |
| Haiku 4.5 | $0.00009 | $0.00151 |
Grade A, and why
reality-inbound 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 yesterday.
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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
REALITY inbound
REALITY hides the server by completing a real TLS 1.3 handshake with a third-party
site (dest) and stealing its certificate for anyone who is not a genuine client.
There is no domain to register and no certificate to renew — but the whole design
rests on picking a good dest, and that choice is not something the panel checks
for you.
Choosing the target
Four properties are mandatory, and scan_reality_target verifies all of them:
TLS 1.3, HTTP/2 in ALPN, X25519 key exchange, and a certificate that chains to a
trusted root. A candidate failing any one of them cannot be used at all.
The rest is judgment the scanner cannot make:
- Not fronted by a CDN with nodes in your users' country. Two separate problems: your server becomes a port-forwarder into that CDN for anyone who probes it, and a censor sees traffic leaving to a foreign IP for a domain that resolves locally — an anomaly that stands out precisely because the domain is popular.
- Not one of the giants.
google.com,microsoft.com,apple.comand friends are the most-analysed SNIs on earth and are usually CDN'd locally. Xray-core agrees loudly enough to have hard-coded it: aserverNamesentry containingappleoricloudlogs "REALITY: Choosing apple, icloud, etc. as the target may get your IP blocked by the GFW" at startup. - Reachable and not itself blocked where the users are. A blocked SNI is worse than no camouflage.
- Low RTT from the panel server, not from you. Every client handshake is
proxied to
dest, so its latency is added to connection setup.scan_reality_targetsranks candidates by feasibility first, then latency. - Plausibly close to the VPS. A site in the same datacenter or AS as the server is both fast and unremarkable.
The trick worth knowing: scan_reality_targets accepts CIDR ranges and discovers
candidates by reading the certificates hosts actually serve. Scanning the
neighbourhood of your own VPS finds local, low-latency, unpopular targets that no
published "good dest" list contains — and published lists are exactly the ones a
censor also reads.
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.
- yesterday First seen · 121 lines · 94 tokens per session scan A 8bc7379b0a5d
reality-inbound is a skill published in the GitHub repository pyworkload/3x-ui-mcp (5 stars, last pushed 4d ago), licensed MIT. It adds 94 tokens to every session and 1,512 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…
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.
event-staffing-ordering
Order W-2 event staff for US/CA events through TempGuru.
chainstack
Work with the Chainstack MCP server to manage blockchain nodes and projects. Use when deploying nodes, migrating to Chainstack from another RPC provider such as QuickNode or Alchemy, checking platform status, searching Chainstack docs, contacting the Chainstack team, or interacting with the Chainstack platform API via…
mcp2cli
Use mcp2cli to connect to, explore, and execute commands against any MCP (Model Context Protocol) server from the terminal. Covers discovery, tool invocation, resource reading, prompt execution, alias creation, config management, background jobs, authentication, and JSON output pipelines.