Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From tldr-swinton
The MCP server is the primary agent interface. 24 tools organized by category.
npx claudepluginhub mistakeknot/interagency-marketplace --plugin tldr-swintonHow this agent operates — its isolation, permissions, and tool access model
Agent reference
tldr-swinton:agents/mcp-serverThe summary Claude sees when deciding whether to delegate to this agent
The MCP server is the primary agent interface. 24 tools organized by category. **Cost ladder** (cheapest first): 1. `extract(file, compact=True)` ~200 tok -- file map (use instead of Read for overview) 2. `structure(project)` ~500 tok -- directory symbols (use instead of Glob + Reads) 3. `context(entry)` ~400 tok -- call graph around symbol (use instead of reading caller files) 4. `diff_context...
Expert Java code reviewer for Spring Boot and Quarkus projects. Automatically detects the framework and applies appropriate review rules. Covers layered architecture, JPA/Panache, MongoDB, security, and concurrency.
Share bugs, ideas, or general feedback.
tldr-code)The MCP server is the primary agent interface. 24 tools organized by category.
Cost ladder (cheapest first):
extract(file, compact=True) ~200 tok -- file map (use instead of Read for overview)structure(project) ~500 tok -- directory symbols (use instead of Glob + Reads)context(entry) ~400 tok -- call graph around symbol (use instead of reading caller files)diff_context(project) ~800 tok -- changed-code context (use instead of git diff + Read)impact(function) ~300 tok -- reverse call graph (use before refactoring)semantic(query) ~300 tok -- meaning-based search (use instead of Grep for concepts)| Category | Tool | Description |
|---|---|---|
| Navigation | tree | File tree listing (paths only, no symbols) |
structure | All symbols across a directory | |
search | Regex search (prefer built-in Grep) | |
extract | Function/class signatures from a file | |
| Context | context | Call graph from a symbol with presets |
diff_context | Git-aware diff context with symbol mapping | |
distill | Compressed prescriptive context for sub-agent handoff | |
delegate | Prioritized retrieval plan for complex tasks | |
| Flow Analysis | cfg | Control flow graph (basic blocks, cyclomatic complexity) |
dfg | Data flow graph (variable def-use chains) | |
slice | Program slice (lines affecting/affected by a given line) | |
| Codebase Analysis | impact | Reverse call graph (find all callers) |
dead | Unreachable code detection (expensive) | |
arch | Architectural layer detection (expensive) | |
calls | Full cross-file call graph (expensive) | |
| Import Analysis | imports | Parse imports from a source file |
importers | Find files importing a given module | |
| Semantic Search | semantic | Meaning-based code search via embeddings |
semantic_index | Build/rebuild semantic index | |
semantic_info | Index metadata (backend, model, count) | |
| Quality | diagnostics | Type checker + linter (pyright + ruff) |
change_impact | Find tests affected by changed files | |
verify_coherence | Cross-file consistency after multi-file edits | |
| Structural | structural_search | AST pattern matching via ast-grep |
| Admin | hotspots | Most-accessed symbols across sessions |
status | Daemon uptime and cache statistics |
Semantic index: Run semantic_index() once before semantic(). Use semantic_info() to check status. Backends: faiss (lighter, Ollama) or colbert (better retrieval, heavier).