opencode-skills-doctor 0.6.2 → 0.7.0

package/README.md

@@ -116,6 +116,7 @@ All commands accept `--json`, `--lang en|zh`, `--budget`, `--root` and `--no-glo
 | **DeepSeek Harness** | `install --client dsh` adds the MCP entry and the `skills-doctor-audit` skill |
 | **Claude Code / Codex** | MCP server (`npx -y opencode-skills-doctor mcp`) |
 | **Grok CLI / Cursor** | their skill roots are scanned natively |
+| **Hermes Agent** | MCP entry in `~/.hermes/config.yaml`; `~/.hermes/skills` is scanned |
 
 ## Why it is safe
 

package/docs/README.zh-CN.md

@@ -339,7 +339,7 @@ Claude Code（项目里的 `.mcp.json`，或你的 MCP 配置）：
 }
 ```
 
-Codex（`~/.codex/config.toml`）：
+Codex（`~/.codex/config.toml`，或 `codex mcp add skills-doctor -- npx -y opencode-skills-doctor mcp`）：
 
 ```toml
 [mcp_servers.skills-doctor]
@@ -347,6 +347,8 @@ command = "npx"
 args = ["-y", "opencode-skills-doctor", "mcp"]
 ```
 
+已用 codex-cli 0.145 验证：`codex mcp list` 显示该 server 为 `enabled`（stdio）。
+
 其他 MCP 客户端：启动 `npx -y opencode-skills-doctor mcp` 并允许 `skills_doctor` 工具即可。
 
 它扫描 `.opencode/skills`、`.claude/skills`、`.agents/skills`（项目和全局），所以其他 agent 用的 skill 也能一起体检。
@@ -412,7 +414,29 @@ skills-doctor check --root .        # 覆盖当前项目的 Grok skill
 skills-doctor check --root $HOME    # 覆盖 ~/.grok/skills 等用户级 skill
 ```
 
-不需要额外安装——Grok 原生加载 `SKILL.md`。如果该 Grok 版本支持 MCP，还可以把上面的 MCP server 接进去，让模型自己调用 `skills_doctor`。
+不需要额外安装——Grok 原生加载 `SKILL.md`。
+
+Grok 1.0+ 也支持 MCP，注册一次即可：
+
+```powershell
+grok mcp add skills-doctor -- npx -y opencode-skills-doctor mcp
+grok mcp doctor skills-doctor   # ✓ server started · handshake OK · 1 tool discovered
+```
+
+Grok 的工具命名是 `<server>__<tool>`，所以调用 `skills_doctor__skills_doctor`；配置写入 `~/.grok/config.toml` 的 `[mcp_servers.skills-doctor]`。
+
+### Hermes Agent（Nous Research）
+
+Hermes 从 `~/.hermes/config.yaml` 读取 MCP server：
+
+```yaml
+mcp_servers:
+  skills_doctor:
+    command: "npx"
+    args: ["-y", "opencode-skills-doctor", "mcp"]
+```
+
+Hermes 的工具命名是 `mcp_<server>_<tool>`，即 `mcp_skills_doctor_skills_doctor`。它的 skill 在 `~/.hermes/skills`（含导入的 `openclaw-imports/`）和项目 `.hermes/skills`，`check` 都会扫描。
 
 ## 检测规则
 

package/docs/REFERENCE.md

@@ -10,7 +10,7 @@ Local scanner, linter, fixer, updater and **OpenCode plugin** for agent skills
 |---|---|
 | **What it is** | Audit, fix and update the skills your coding agent loads |
 | **Interfaces** | CLI · OpenCode plugin (1.x / 2.x) · MCP server (Claude Code, Codex, DSH, …) · Grok CLI compatible |
-| **Skill sources** | `.opencode/skills`, `.claude/skills`, `.agents/skills`, `.grok/skills`, `.cursor/skills`, `opencode.json(c)` entries, HTTP catalogs |
+| **Skill sources** | `.opencode/skills`, `.claude/skills`, `.agents/skills`, `.grok/skills`, `.cursor/skills`, `.hermes/skills`, `opencode.json(c)` entries, HTTP catalogs |
 | **What it finds** | shadowed IDs, duplicates, long descriptions, oversized bodies, per-step token budget, broken file references, invalid frontmatter, empty bodies, stale skills |
 | **What it fixes** | archives shadowed definitions (reversible), hides low-frequency skills, syncs catalogs with drift detection |
 | **Reports** | terminal, JSON, or a self-contained HTML report (`--html [file] --open`) |
@@ -322,7 +322,7 @@ Claude Code (`.mcp.json` in the project, or your MCP config):
 }
 ```
 
-Codex (`~/.codex/config.toml`):
+Codex (`~/.codex/config.toml`, or `codex mcp add skills-doctor -- npx -y opencode-skills-doctor mcp`):
 
 ```toml
 [mcp_servers.skills-doctor]
@@ -330,6 +330,8 @@ command = "npx"
 args = ["-y", "opencode-skills-doctor", "mcp"]
 ```
 
+Verified with codex-cli 0.145: `codex mcp list` shows the server as `enabled` (stdio).
+
 Any other MCP client: run `npx -y opencode-skills-doctor mcp` and allow the `skills_doctor` tool.
 
 It scans `.opencode/skills`, `.claude/skills` and `.agents/skills` (project and global), so it can audit skills used by other agents too.
@@ -395,7 +397,29 @@ skills-doctor check --root .        # covers your Grok project skills
 skills-doctor check --root $HOME    # covers ~/.grok/skills and other user-level skills
 ```
 
-No install step is needed — Grok loads `SKILL.md` packages natively. If your Grok build has MCP enabled, you can additionally register the MCP server above to let the model call `skills_doctor` itself.
+No install step is needed — Grok loads `SKILL.md` packages natively.
+
+Grok 1.0+ also speaks MCP. Register the tool once:
+
+```powershell
+grok mcp add skills-doctor -- npx -y opencode-skills-doctor mcp
+grok mcp doctor skills-doctor   # ✓ server started · handshake OK · 1 tool discovered
+```
+
+Grok namespaces tools as `<server>__<tool>`, so call `skills_doctor__skills_doctor`. The config lands in `~/.grok/config.toml` as `[mcp_servers.skills-doctor]`.
+
+### Hermes Agent (Nous Research)
+
+Hermes reads MCP servers from `~/.hermes/config.yaml`:
+
+```yaml
+mcp_servers:
+  skills_doctor:
+    command: "npx"
+    args: ["-y", "opencode-skills-doctor", "mcp"]
+```
+
+Hermes names tools `mcp_<server>_<tool>`, so the tool is `mcp_skills_doctor_skills_doctor`. Its skills live in `~/.hermes/skills` (including imported `openclaw-imports/`) and project `.hermes/skills`; `check` scans both.
 
 ## Rules
 
@@ -440,9 +464,11 @@ No install step is needed — Grok loads `SKILL.md` packages natively. If your G
 6. project `.cursor/skills`
 7. `~/.grok/skills`
 8. project `.grok/skills`
-9. `~/.config/opencode/skills`
-10. project `.opencode/skills`
-11. the `skills` array in `opencode.json(c)` (2.x) or `skills.paths` (1.x); relative paths resolve from the working directory; supports `~/`, absolute paths and HTTP catalogs
+9. `~/.hermes/skills`
+10. project `.hermes/skills`
+11. `~/.config/opencode/skills`
+12. project `.opencode/skills`
+13. the `skills` array in `opencode.json(c)` (2.x) or `skills.paths` (1.x); relative paths resolve from the working directory; supports `~/`, absolute paths and HTTP catalogs
 
 ## JSON output
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-skills-doctor",
-  "version": "0.6.2",
+  "version": "0.7.0",
   "description": "Audit, fix and update agent skills: token cost, broken references, shadowed duplicates, HTML reports, catalog sync. CLI, OpenCode plugin, MCP server; works with OpenCode, Claude Code, Codex, Grok CLI, Cursor and DeepSeek Harness.",
   "type": "module",
   "exports": {

package/src/discover.mjs

@@ -120,9 +120,11 @@ function isUnder(parent, child) {
  *  6. 项目 .cursor/skills
  *  7. ~/.grok/skills
  *  8. 项目 .grok/skills
- *  9. ~/.config/opencode/skills
- * 10. 项目 .opencode/skills（项目根 → cwd）
- * 11. opencode.json(c) 里的 skills 数组 / skills.paths
+ *  9. ~/.hermes/skills（Hermes Agent）
+ * 10. 项目 .hermes/skills
+ * 11. ~/.config/opencode/skills
+ * 12. 项目 .opencode/skills（项目根 → cwd）
+ * 13. opencode.json(c) 里的 skills 数组 / skills.paths
  */
 export function discoverSources({ cwd = process.cwd(), home = os.homedir(), includeGlobal = true } = {}) {
   const projectRoot = findProjectRoot(cwd) ?? path.resolve(cwd)
@@ -151,6 +153,9 @@ export function discoverSources({ cwd = process.cwd(), home = os.homedir(), incl
   if (includeGlobal) addSource("grok-global", "global", path.join(home, ".grok", "skills"))
   for (const dir of projectChain) addSource("grok-project", "project", path.join(dir, ".grok", "skills"))
 
+  if (includeGlobal) addSource("hermes-global", "global", path.join(home, ".hermes", "skills"))
+  for (const dir of projectChain) addSource("hermes-project", "project", path.join(dir, ".hermes", "skills"))
+
   if (includeGlobal) addSource("opencode-global", "global", path.join(home, ".config", "opencode", "skills"))
   for (const dir of projectChain) addSource("opencode-project", "project", path.join(dir, ".opencode", "skills"))