Search Router

This is a routing guide that tells Claude which search tool to use for different query types to save tokens. The key insight is using TLDR search as the default for code exploration instead of grep, since it gives you 95% token savings. It breaks down when to use AST-grep for structural patterns, TLDR semantic for conceptual questions, and grep only as a last resort for literal text. The decision tree is solid and the token comparison table makes the efficiency gains concrete. If you're working with Claude on a large codebase and burning through context with inefficient searches, this gives you a practical playbook for staying within budget.

skills.sh installs: 487
GitHub stars: 3.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add parcadei/continuous-claude-v3 --skill search-router --agent claude-code

Installs into .claude/skills of the current project.

Files

Search Tool Router

Use the most token-efficient search tool for each query type.

When to Use

Searching for code patterns
Finding where something is implemented
Looking for specific identifiers
Understanding how code works

Decision Tree

Query Type?
├── CODE EXPLORATION (symbols, call chains, data flow)
│   → TLDR Search - 95% token savings
│   DEFAULT FOR ALL CODE SEARCH - use instead of Grep
│   Examples: "spawn_agent", "DataPoller", "redis usage"
│   Command: tldr search "query" .
│
├── STRUCTURAL (AST patterns)
│   → AST-grep (/ast-grep-find) - ~50 tokens output
│   Examples: "def foo", "class Bar", "import X", "@decorator"
│
├── SEMANTIC (conceptual questions)
│   → TLDR Semantic - 5-layer embeddings (P6)
│   Examples: "how does auth work", "find error handling patterns"
│   Command: tldr semantic search "query"
│
├── LITERAL (exact text, regex)
│   → Grep tool - LAST RESORT
│   Only when TLDR/AST-grep don't apply
│   Examples: error messages, config values, non-code text
│
└── FULL CONTEXT (need complete understanding)
    → Read tool - 1500+ tokens
    Last resort after finding the right file