claude-code-skills MCP server
Plugin suite + bundled MCP servers for Claude Code. Full delivery lifecycle: Agile pipeline with multi-model AI review, project bootstrap, documentation generation, codebase audits, performance optimization, community workflows. Includes hex-line (hash-verified editing), hex-graph (code knowledge graph), and hex-ssh (remote SSH) MCP servers.
How to read this: tool names here are observed from a live tools/list handshake. The Risk label is a heuristic inferred from the tool name (write/destructive verbs), not from executing the tool — a conservative guess, not a verified capability. We never escalate risk from a description. Found one that's wrong? Tell us — we fix on report.
| Tool | Risk | Side effects | Approval |
|---|---|---|---|
| verify Check if held checksums are still valid without rereading. Use before delayed or mixed-tool follow-up edits; returns canonical status, next_action, and reread guidance. | read | false | unknown |
| write_file Create a new file or overwrite existing. Creates parent dirs. For existing files prefer edit_file (shows diff, verifies hashes). | write | true | unknown |
| edit_file Apply hash-verified partial edits to one file. Batch ALL hunks for the same file into ONE call via the `edits` array -- separate sequential edit_file calls on one file go stale and conflict (the most common edit failure). Carry base_revision only for a genuinely later follow-up after the file changed. Anchors accept tag.N, a bare line number, or unique line content (auto-resolved); range_checksum accepts "auto" to compute it for the current range. Preserves existing line endings and trailing-newline shape; conservative conflicts return retry helpers. boundary_mode=inclusive deletes the anchor lines themselves; new_text must close any delimiter whose opening falls inside the replaced range. | write | true | unknown |
| read_file Read file with progressive disclosure. Default: minimal plain partial read for discovery; enable edit-ready metadata explicitly when preparing a verified edit. | read | false | unknown |
| changes Semantic diff against git ref (default: HEAD). Returns canonical status, summary, next_action, changed symbols, and graph-backed risk hints when available. | read | false | unknown |
| grep_search Search file contents with ripgrep. Default: summary-first discovery output; use `content` with `edit_ready=true` when you need verified edit hunks. | read | false | unknown |
| outline AST-based structural outline with hash anchors for direct edit_file usage. Supports JavaScript/TypeScript, Python, C#, and PHP code files plus markdown headings (.md/.mdx, fence-aware). | unknown | unknown | unknown |
| inspect_path Inspect a file or directory path. Files return compact metadata; directories return a gitignore-aware tree or pattern matches. | read | false | unknown |
| bulk_replace Search-and-replace text across multiple files inside an explicit root path. Use for renames and refactors when the project scope is known. | read | false | unknown |
- repohttps://github.com/levnikolaevich/claude-code-skills
- homepagehttps://levnikolaevich.github.io/claude-code-skills/
- licenseMIT
- adoption538 stars · 75 forks
The access this server can exercise, inferred from its verified tools — not a declared OAuth scope.
Add the “as seen on MCPExplorer” badge to your README. 
This is one server. A loadout combines the right servers, governance, and proven plays for a whole job — assembled deliberately, not tool-dumped.
Explore loadouts →