OpeItcLoc03/claude-skills
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(retire): drop using-synology-ops skill — NAS decommissioned

Browse Source 
Synology NAS permanently retired 2026-05-25. Removing dead-code:

- skills/using-synology-ops/ — source deleted
- dist-hermes/mcp/using-synology-ops/ — built artifact deleted
- hermes/mapping.yaml — entry removed
- skills/using-vds-ops/SKILL.md — stripped NAS-disambiguation
  clause from description + "Mirror of using-synology-ops" line
  from body; bumped 0.1.0 → 0.1.1 PATCH (wording cleanup post-
  retirement, no capability change)

Out of repo (user-config side, manual follow-up):
- ~/.claude/skills/using-synology-ops/ install dir
- ~/.claude.json mcp.synology-ops server entry (dead URL +
  bearer token, must drop)
- C:/Users/vitya/projects/synology-ops-mcp/ source repo —
  separate decision, out of this scope
This commit is contained in:
vitya
2026-05-25 07:33:29 +03:00
parent c62d6c3391
commit 3f8262b98e
4 changed files with 3 additions and 165 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
skills/using-vds-ops/SKILL.md
View File

@@ -1,12 +1,12 @@
---
name: using-vds-ops
version: 0.1.0
description: "Use when diagnosing docker containers on the Rusonyx VDS — the `vds-ops` HTTP MCP server (https://opsmcp.vds.kzntsv.site/mcp) exposes 4 read-only tools — `ops.docker.ps`, `ops.docker.logs`, `ops.docker.inspect`, `ops.docker.stats` — all routed through tecnativa `vds-docker-proxy-ro` with `POST=0` (write-ops return 403). Triggers on VDS container names (`traefik`, `portainer`, `gitea`, `verdaccio`, `registry`, `joxit-ui`, `ntfy`, `postgres`, `mariadb`, `mongo`, `redis`, `vds-ops-mcp`, `vds-docker-proxy-ro`), on VDS-domain words («на VDS», «VDS», «vds.kzntsv.site», «vds-ops», «opsmcp.vds», «Rusonyx», «облачный VDS»), or on incident phrases («падает», «restart-loop», «не стартует», «unhealthy», «посмотри логи», «git.kzntsv не открывается», «verdaccio лежит», «registry медленно», «ntfy не приходит») when VDS-context. Read-only by design — no per-session grant needed. Disambiguation: `traefik` exists on both VDS and NAS — ask «на VDS или на NAS?» before tool call when ambiguous. Skip for non-VDS hosts (NAS `modulair-*` → using-synology-ops), write-ops, mentions without incident signal."
version: 0.1.1
description: "Use when diagnosing docker containers on the Rusonyx VDS — the `vds-ops` HTTP MCP server (https://opsmcp.vds.kzntsv.site/mcp) exposes 4 read-only tools — `ops.docker.ps`, `ops.docker.logs`, `ops.docker.inspect`, `ops.docker.stats` — all routed through tecnativa `vds-docker-proxy-ro` with `POST=0` (write-ops return 403). Triggers on VDS container names (`traefik`, `portainer`, `gitea`, `verdaccio`, `registry`, `joxit-ui`, `ntfy`, `postgres`, `mariadb`, `mongo`, `redis`, `vds-ops-mcp`, `vds-docker-proxy-ro`), on VDS-domain words («на VDS», «VDS», «vds.kzntsv.site», «vds-ops», «opsmcp.vds», «Rusonyx», «облачный VDS»), or on incident phrases («падает», «restart-loop», «не стартует», «unhealthy», «посмотри логи», «git.kzntsv не открывается», «verdaccio лежит», «registry медленно», «ntfy не приходит») when VDS-context. Read-only by design — no per-session grant needed. Skip for write-ops and mentions without incident signal."
---
# using-vds-ops
Read-only diagnostic access to docker containers on the Rusonyx VDS via the `vds-ops` HTTP MCP server. Mirror of `using-synology-ops` for the VDS host. Skeleton — body to be filled in a follow-up pass.
Read-only diagnostic access to docker containers on the Rusonyx VDS via the `vds-ops` HTTP MCP server. Skeleton — body to be filled in a follow-up pass.
## When to use