Cross-Repo Workflow comparison

cross_repo_search

Search symbols across all indexed repos simultaneously. Supports repo_pattern filtering.

−95%
Token reduction
N× grep per repo
Native baseline
1 vs 3
Calls (CS vs native)
800 vs 16,900
Tokens (CS vs native)

What it does

Searches for symbols across all indexed repositories simultaneously. Supports repo_pattern filtering to scope to a subset.

Benchmark

ApproachTokensCalls
Native (3 separate grep calls)16,9173
cross_repo_search8031
Reduction−95%−67%

The deeper value

When a utility function is used across 5 repos and you’re about to change its signature, cross_repo_search tells you where it’s defined — across all repos — in one call. Native tooling requires separate grep runs and manual result aggregation.

Benchmark note

This benchmark compares CodeSift against the closest practical native workflow an agent would use for the same task. For some tools, that baseline is a direct shell equivalent such as rg or find. For AST-aware, graph-aware, and LSP-backed tools, the baseline is a multi-step workflow rather than a strictly identical command. Results should be read as agent-workflow comparisons: token cost, call count, and practical context efficiency.