Initial commit: Hermes skill tproxy-web-proxy

SKILL.md

@@ -0,0 +1,108 @@
+---
+name: tproxy-web-proxy
+description: >-
+  Operate tproxy-server (secrets, readyz, metrics).
+tags:
+  - telegram
+  - proxy
+  - mtproto
+  - web-proxy
+  - systemd
+  - prometheus
+category: devops
+---
+
+# tproxy-web-proxy — Telegram WEB Proxy (telegramdesktop/tproxy-server)
+
+## When to Use
+
+- Add / revoke client secrets (profiles) on a deployed tproxy-server
+  («добавь ещё один секрет», «убери доступ другу»)
+- Check proxy health (/readyz, /healthz) or answer what an endpoint returns
+- Wire /metrics into the Prometheus stack (/opt/monitoring)
+- Troubleshoot the service (Caddy → tproxy-server → MTProxy chain)
+- Deploy tproxy-server on a fresh VPS
+
+## Architecture (deployed on vps03, 2026-08)
+
+```
+Telegram Desktop (WEB Proxy) --HTTPS 443--> Caddy (vps03.nixg.ru)
+  --> tproxy-server :8080 (loopback) --> MTProxy :2398 (loopback)
+```
+
+Key files (Debian):
+- `/etc/tproxy-server/config.json` — main config (`listen` 127.0.0.1:8080,
+  `admin_listen` 127.0.0.1:8081, `limits.max_profiles` default 32)
+- `/etc/tproxy-server/profiles.json` — ARRAY of client profiles (secrets),
+  perms `600 root:tproxy`
+- `/etc/systemd/system/tproxy-server.service` —
+  `ExecStart=/usr/local/bin/tproxy-server -config ...`,
+  `LoadCredential=profiles.json:/etc/tproxy-server/profiles.json`
+- `/etc/tproxy-server/firewall.nft` — nft rules
+
+## Multiple secrets (profiles)
+
+profiles.json is an ARRAY → many secrets supported (limit `max_profiles: 32`).
+Each profile points to the same MTProxy backend unless overridden — one
+MTProxy, many access keys.
+
+```json
+{"profiles":[
+  {"name":"default","secret":"<32-hex>","backend":"127.0.0.1:2398"},
+  {"name":"rahuba","secret":"<32-hex>","backend":"127.0.0.1:2398"}
+]}
+```
+
+Secret format: 16 bytes = 32 hex (plain) OR 17 bytes with `dd` prefix
+(fake-TLS mode). `openssl rand -hex 16` → plain 32-hex.
+
+Add a profile:
+1. Edit `/etc/tproxy-server/profiles.json` (python3 json.dump for safety)
+2. `chmod 600` + `chown root:tproxy` — systemd LoadCredential rejects
+   group/other-readable files, unit fails to start otherwise
+3. `systemctl restart tproxy-server`
+4. Verify: `journalctl -u tproxy-server | grep 'event=started'` shows
+   `profiles=N` (N = profile count); `curl :8081/readyz` → 200
+
+Revoke: remove the profile object + restart.
+
+## Admin endpoints (`admin_listen`, loopback only)
+
+- `/healthz` — process alive, always 200 `ok`; does NOT check backend
+- `/readyz` — TCP `net.DialTimeout` to EVERY profile's backend (5s);
+  all ok → 200 `ready`; any dead → 503 `backend unavailable`.
+  NOTE: one profile with a dead backend makes readyz fail for ALL.
+- `/metrics` — Prometheus text format (13 counters; full list, scrape config
+  and alert ideas in `references/tproxy-monitoring.md`)
+- `/debug/pprof/*` — only when `enable_pprof: true`
+
+## Pitfalls
+
+1. **Install: `go test ./...` fails under root** —
+   `TestLoadAcceptsSystemdCredentialReadPermissions` («group/other-readable
+   profiles file outside a credential directory was accepted»). Run the
+   build/tests as a non-root user, or skip that single test. Do NOT change
+   `loadProfiles` semantics in upstream code.
+2. **profiles.json permissions** — must not be group/other readable/writable
+   (use `600 root:tproxy`). This is exactly what the config unit test enforces.
+3. **readyz is per-backend** — a dead backend in any profile → 503 for
+   everyone.
+4. **Admin endpoints are loopback-only** — to expose /metrics to a remote
+   Prometheus, open the port in nft restricted to the monitor's source IP
+   (e.g. bigbox public IP), never 0.0.0.0.
+5. **Restart required after every profiles.json change** — profiles are
+   loaded once at start via LoadCredential.
+6. **Keep secrets out of shell history** — pass via file `$(cat ...)`, never
+   inline; tproxy logs never print secrets.
+
+## Monitoring
+
+Full metrics table, readyz semantics, prometheus.yml job and alert ideas:
+`references/tproxy-monitoring.md`. Monitoring task lives in
+`/opt/monitoring/PLAN.md` («Этап 6. Метрики tproxy-server (vps03)»).
+
+## Related Skills
+
+- `mtproto-proxy` — classic MTProto proxy via seriyps docker image (DIFFERENT
+  product; tproxy-server is the official browser-based WEB proxy)
+- `networking-proxy` — umbrella for the XRay/AmneziaWG/Shadowsocks stack

references/tproxy-monitoring.md

@@ -0,0 +1,85 @@
+# tproxy-server monitoring reference
+
+Admin HTTP endpoints (bound to `127.0.0.1:8081` = `admin_listen`, loopback only):
+
+| Endpoint | Returns |
+|---|---|
+| `/healthz` | always `200 ok` while the process runs; does NOT touch backends |
+| `/readyz` | `200 ready` if TCP dial succeeds to EVERY profile backend; `503 backend unavailable` if any backend dial fails (timeout = `timeouts.backend_dial`, default 5s) |
+| `/metrics` | Prometheus text format (`Content-Type: text/plain; version=0.0.4`), 13 counters |
+| `/debug/pprof/*` | only when `config.enable_pprof: true` |
+
+## Metrics (from serveMetrics, internal/server/server.go)
+
+```
+tproxy_sessions_live              # currently active sessions
+tproxy_streams_live               # currently active streams
+tproxy_backend_dials_in_flight    # in-flight dials to MTProxy (127.0.0.1:2398)
+tproxy_pending_bytes              # queued payload waiting to be relayed
+tproxy_pending_items              # queued items
+tproxy_sessions_created_total     # cumulative
+tproxy_sessions_closed_total      # cumulative
+tproxy_streams_opened_total       # cumulative
+tproxy_streams_rejected_total     # cumulative
+tproxy_backend_dial_failures_total  # cumulative — growth = MTProxy down/unreachable
+tproxy_bytes_up_total             # cumulative, client→proxy→backend
+tproxy_bytes_down_total           # cumulative, backend→client
+tproxy_limit_hits_total           # cumulative, rate-limit / capacity hits
+```
+
+Note: live values come from `manager.Capacity()` (current), `_total` counters
+from `manager.Metrics()` (cumulative). No labels/help strings emitted — plain
+`name value` lines.
+
+## Prometheus scrape config (prometheus runs network_mode: host)
+
+```yaml
+- job_name: 'tproxy'
+  static_configs:
+    - targets: ['77.67.89.154:8081']   # vps03 public IP; port must be opened
+      labels:
+        host: vps03
+        service: tproxy
+```
+
+## Exposing 8081 to the monitor (vps03, nft)
+
+Endpoint listens on loopback only — open for ONE source IP, e.g. bigbox:
+
+```bash
+nft add rule inet filter input ip saddr <BIGBOX_IP> tcp dport 8081 accept
+systemctl restart tproxy-server   # or nft reload
+```
+
+Never publish to 0.0.0.0; the endpoint has no auth.
+
+## Alert ideas
+
+- `tproxy_backend_dial_failures_total` increases → MTProxy unreachable
+- `/readyz` non-200 → chain broken (but remember: ONE dead profile backend
+  fails readyz for everyone)
+- `tproxy_limit_hits_total` increases → abuse / capacity exceeded
+- `up{job="tproxy"} == 0` → exporter itself down
+
+## Monitoring task location
+
+`/opt/monitoring/PLAN.md` — «Этап 6. Метрики tproxy-server (vps03)» (written
+2026-08-31),  git repo `gitverse.ru:kpa39l/monitoring.git`, gitea mirror.
+
+## Verified live sample (vps03, 2026-08-31, idle-ish)
+
+```
+tproxy_sessions_live 1
+tproxy_streams_live 5
+tproxy_backend_dials_in_flight 0
+tproxy_pending_bytes 0
+tproxy_pending_items 0
+tproxy_sessions_created_total 1
+tproxy_sessions_closed_total 0
+tproxy_streams_opened_total 15
+tproxy_streams_rejected_total 0
+tproxy_backend_dial_failures_total 0
+tproxy_bytes_up_total 166853
+tproxy_bytes_down_total 5076719
+tproxy_limit_hits_total 0
+```