Discovers, ranks, and composes compiler-verified repository context bundles — maximizing coding-agent task success while eliminating 90% of token bloat under strict economic budget ceilings.
npx siftrcode context "task prompt"
Select your coding agent to view instant setup commands.
Registers SiftrCode's Stdio MCP server to prune code before Claude reads it.
siftr context "task description"
How SiftrCode V2 delivers 100% recall on critical edit targets while compressing token payloads by up to 90%.
Blends Exact symbol matches, BM25 text retrieval, Stack Trace log parsing, Semantic AST Context Graph, and Git Co-Change intelligence to never miss relevant code.
Extracts deterministic feature vectors with strict FeatureCutoff timestamps, strictly eliminating data leakage and lookahead contamination.
Scores candidates with fully inspectable component breakdowns. Applies deterministic penalties to lockfiles, generated code, and unrelated tests.
Assembles bundles with diminishing marginal returns. Maximizes task evidence coverage while suppressing same-file symbol redundancy and dead weight.
Protects direct edit targets at FULL/BODY resolution. Degrades surrounding modules into compiler-verified AST skeletons, and non-code files safely.
Strictly enforces token caps (e.g. 16,000 max) and economic cost ceilings (e.g. $0.05/turn) with precise pricing models for Sonnet 3.7 and GPT-4o.
The math is easy. You could be saving thousands in agent fees*
Method bodies dissolved • 100% signatures & types preserved.
Saves $0.35/turn • $4,560 net annual savings per engineer.
$
npx siftrcode audit
* Analysis for Claude 3.7 Sonnet ($3.00/MTok in) based on 50 agent turns / day.
Free open-source CLI for individuals. Paid caching for teams.
siftr context)