lci

Plugin lci from marketplace standardbeagle-tools in standardbeagle/standardbeagle-tools. Commands read from the repo's manifest — 4 days ago.

Skills (10)

• lci-agnt-companion

  Use when lci is installed and user needs dev server, reverse proxy, browser debugging, screenshots, or accessibility/performance/security audits of running page - points at sibling agnt plugin which lci does not duplicate. lci搜索代碼；運行代碼、觀其所為則用agnt。 Use when: dev server, browser errors, screenshots, a

• lci-commit-all

  Orchestrate full WIP preparation and auto-commit by loading project config from memory and dispatching dedicated agents. 統籌WIP準備，自動提交：從記憶加載配置，分派專屬代理。 Use when: wrapping up uncommitted work, single-command test-doc-commit, end of work session.

• lci-context-handoff

  Save and load code context manifests for agent handoff and session continuity. 保存/加載代碼上下文清單供代理交接。 Use when: handing off to subagent, saving investigation state, building incremental context, multi-session continuity.

• lci-explore-codebase

  Understand codebase architecture, structure, and patterns using LCI code intelligence. 以LCI探索代碼庫架構結構。 Use when: first look at unfamiliar codebase, understanding module layout, finding entry points, assessing complexity.

• lci-force-lci

  Toggle force-LCI mode to block standard search tools and require LCI. 強制啟用LCI模式，封鎖Grep/Glob。 Use when: forcing all searches through LCI, benchmarking LCI, disabling force mode.

• lci-mcp-status

  Check LCI MCP server registration status across slop-mcp and standard configurations. 查LCI伺服器登錄狀態。 Use when: checking if lci is registered, diagnosing connection errors, verifying index status.

• lci-pre-commit-review

  Analyze staged or uncommitted code changes for quality issues before committing. 提交前以LCI分析暫存更改之質量問題。 Use when: pre-commit quality check, PR review, finding duplicates, checking naming conventions, analyzing complexity.

• lci-search-code

  Find code symbols, patterns, implementations, and files using LCI semantic search. 以LCI語義搜索查找符號、模式、實現、文件。 Use when: finding function/class/type definitions, searching symbol usages, finding files by path, filtering by language or directory.

• lci-setup-mcp

  Install lci MCP server with intelligent detection - uses the local binary, installing it first when missing. 智能安裝LCI MCP伺服器，無本地二進位則先裝之。 Use when: first-time lci setup, registering with slop-mcp, configuring mcp.json.

• lci-trace-symbol

  Trace function call hierarchies, dependencies, side effects, and symbol relationships using LCI. 以LCI追蹤函數調用層次、依賴、副作用、符號關係。 Use when: understanding function before modifying, finding all callers before refactoring, checking side effects, tracing call chain for debugging.

Subagents (6)

• code-quality

  Performs code quality analysis using LCI, runs linters and formatters, removes debug artifacts, and fixes all findings. 以LCI分析代碼質量，運行lint/format，移除調試產物，修復所有發現。 Use when: ensuring code meets quality standards, linting and formatting changes, checking for quality issues before commit.

• context-gatherer

  Gathers all work-in-progress context from git, tasks, plans, and LCI analysis. Returns a structured change summary for other agents to consume. 從git、任務、計劃、LCI分析收集所有WIP上下文，返回供其他代理使用之結構化更改摘要。 Use when: collecting context before dispatching agents, summarizing changes for commit, gathering WIP informat

• doc-updater

  Updates internal project documentation — CHANGELOG, inline docs, plan files — to reflect current changes. Only updates doc types that already exist in the project. 更新內部文檔：CHANGELOG、內聯文檔、計劃文件，反映當前更改；僅更新項目中已存在之文檔類型。 Use when: updating docs after code changes, adding changelog entry, tracking plan comp

• marketing-seo

  Updates public-facing documentation (README, package metadata) with clear, benefit-oriented language and SEO-aware content. Skips if changes are purely internal. 更新公開文檔（README、包元數據）為清晰利益導向語言；純內部更改則跳過。 Use when: updating README for new features, improving package description, making docs more discove

• tdd-refactor

  Antagonistic TDD refactoring agent — lci semantic search + code intelligence to find testability issues, hidden dependencies, side effects; guide aggressive refactor that eliminates mocking. 對抗性TDD重構，消除mock需求。 Use when: analyzing code for testability, understanding dependencies, finding refactoring

• test-fixer

  Runs the full test suite, fixes all failures (including pre-existing), and adds missing test coverage for changed code. Uses LCI to find uncovered code paths. 運行全部測試套件，修復所有失敗（含預存在），為已更改代碼添加缺失覆蓋；以LCI查找未覆蓋代碼路徑。 Use when: fixing broken tests, adding coverage for changes, running and fixing all tests.

Hooks (1)

• PreToolUse

  node "${CLAUDE_PLUGIN_ROOT}/scripts/block-search-tools.js"

Similar plugins