Score breakdown
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for developers running AI coding agents (Claude Code, Codex, Cursor, Aider, OpenHands) who need a faster, token-cheaper, agent-native replacement for `grep` / `cat` / `read` / `ls` inside the agent session: a Rust single-binary CLI with hybrid BM25 + semantic search, Tree-sitter AST chunking, a dependency graph, and a `digest` pipeline for build/CI output; for users who want to drop the age
Who should use it
Who should skip it
Skip johunsang/semble_rs if the source repository or demo is inactive, unmaintained, or no longer matches the description shown here.
About this signal
johunsang/semble_rs is tracked by RepoRadar as a rust bm25 + semantic code search in the MIT Rust single-binary code search CLI (BM25 + M section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Silver tier and easy setup difficulty. johunsang/semble_rs leads on novelty (9.0) and workflow potential (9.0); its lowest signal is momentum (6.0), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.
How this item is evaluated
RepoRadar assigned johunsang/semble_rs a composite score of 7.9 out of 10, placing it in the Silver tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 134.0 and never affects the composite score or tier. The risk label of 'low' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.
Risk explanation
**License is MIT per `Cargo.toml` `license = "MIT"` field, not a top-level LICENSE file.** The repo does NOT ship a plain `LICENSE` at the root — the canonical license declaration is in `Cargo.toml` (`license = "MIT"`) which is the cycle 140 SPDX-OR / Cargo.toml-license-field verification path; the README also carries a `[]` badge. This is MIT-licensed by both metadata signals, but a tooling chain that expects a top-level LICENSE file may report 'no license file' until it parses Cargo.toml. The cycle 140 doctrine confirms Cargo.toml `license = "..."` is the source of truth for Rust projects; **First-run downloads a 60 MB embedder from HuggingFace.** The default embedder `minishlab/potion-code-16M` is downloaded from HuggingFace on first run (cached thereafter); users on offline machines or behind strict egress firewalls should pre-download the embedder, override with `--model <local-path>` pointing at a previously-cached embedder, or set `SEMBLE_MODEL_PATH` to a known location; **Token-reduction numbers are measured on the project's own repos and a 6,693-file Python backend.** The README's headline numbers (747× on `tree`, 64× on the Python backend, 47% on `--outline`, 98.9% on `digest`) are real measurements, not synthetic — but the magnitude of the reduction depends on the project's structure (projects dominated by `target/`, `node_modules/`, `.git/` directories will see the biggest `tree` reductions; projects dominated by source files will see smaller reductions); users should run `semble_rs tree` once on their own repo and compare against `ls -R | wc -c` to set their own baseline before claiming token savings.