User repository on GitHub
Experimental semantic line search with TypeSafe Jev via OpenRouter. Python CLI with no runtime dependencies.
- Forks
- 1
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Practices & Patterns
Jev Search — larguesa
Line-oriented semantic search that adds a Jev meaning pass over your own shortlisted files next to exact search. Real search is the default with a key, while dry-run previews the payload offline with no network or credentials.
- Stars
- 8
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- larguesa
- Last verified
- Oct 2026
This is one of several Jev listings for the same project. The canonical entry is Jev Search — superagents-lab.
About this project
Find by meaning what keywords miss. Complementary semantic search for agents exploring documents and knowledge bases alongside their existing tools. Exact search finds names, symbols and literal references. jev-search adds another perspective: whether a line expresses the intent you are looking for, even with different wording. Keep both, then read the surrounding context. Repository documentation: explore design intentions and trade-offs, then inspect the actual source with your usual tools.
Quoted from the project's README. Read the full README.
Used inJev for Software & DevTools
- Languages
- Python
Highlights
- Complements exact search by judging whether lines express the query intent in other words.
- Evaluates lines in local regular files, with UNC paths, streams, and reparse points rejected.
- Two native decision backends with explicit provider, model, and dedicated key configuration.
- Offline dry-run prints candidates and payload; real search needs authorized scope and budget.
Watch out
Small curated experiments only, not a general accuracy claim; misses and false positives remain and ranking is opt-in. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
Lexical search finds names, symbols, and literal references; this adds whether a line expresses the requested intent in different wording. The workflow is to orient with exact search, shortlist bounded passages, apply this as another pass, then read the original context.
In six simulated tasks over selected real text, lexical plus Jev recovered 29 of 34 labeled passages versus 22 of 34 for lexical alone, with no extra false positives. The digest stresses this is small and manually curated, and adjacent hits might have been found by reading context.
Use Linux or Windows with Python 3.11 plus uv or pipx; macOS is not supported or tested. Clone main, review it, record the revision, install from that same revision, and set JEV_SEARCH_API_KEY for the selected provider with provider-side spending limits.
Ranked results only sort existing matches by descending score with stable ties, and top-k limits that extra view. No extra inference runs, the full JSON gets larger, a shuffled baseline beat ranking in the deeper study, and the default complete results stay preserved.
OpenRouter at its decisions endpoint by default, or TypeSafe directly; set JEV_SEARCH_API_KEY with --provider openrouter|typesafe, and keys are not interchangeable or auto-inherited from other provider variables.
Linux or Windows, Python 3.11+, with uv or pipx; macOS is not supported or tested, and only local regular files in trusted directories are accepted, with UNC and reparse points rejected.
Add --dry-run to skip the network, credentials and meaning evaluation; it prints JSON with selected lines and the request payload, while real search is the default and may incur charges.
More Jev resources like this
7GitHub stars
Four read-only MCP tools that rank files, cut a file down to the relevant part and pick from long lists, measured on a frozen SWE-bench Lite split.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
Bentlybro/siftrPython
Bentlybro/siftrPython4GitHub stars
Python library and CLI that moves bulk semantic judgments — ticket triage, stale-file tagging, bug or question routing — out of the agent's context into typed Jev questions, with a measured comparison against an in-context control.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
ddfeyes/jev-modePython
ddfeyes/jev-modePython4GitHub stars
Framework-agnostic guard layer that puts Jev in front of an agent's tool calls: a Choice risk tier, a Noul destructiveness check and a Score blast radius decide allow, deny or ask.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
lgy1027/jevshieldPython
lgy1027/jevshieldPython2GitHub stars
Runs Jev as an LLM router over two public benchmarks, with a no-Jev ablation and RouterArena's own scorer so the numbers sit next to the public leaderboard.
Jev Practices & Patterns#community#jevbest#catalog-import
1GitHub stars
Runnable benchmark putting Jev, jev-preview and Claude Sonnet 5 through one real agent task — classifying a tool call as readonly, destructive, privileged or exfiltration — on the same 60 labelled cases, with raw per-call results committed.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
themsquared/jev-benchmarkPython
themsquared/jev-benchmarkPython1GitHub stars
pydecide is one Python client for typed System One decisions — Choice, Score and Noul — over hosted Jev, OpenRouter's Decisions endpoint, local laya on PyTorch or MLX, a sentence-transformers CrossEncoder, or a JSON-prompted LLM fallback.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
gopaljigaur/decidePython
gopaljigaur/decidePythonFrom the Jev guides
Original write-ups on what this entry is for.
Most Jev accuracy problems are state problems. What belongs in a state, what is a pre-made judgment to leave out, and why naming the slice you ask about matters more than the wording.
Jev Practices & Patterns#state#design#primitives
Read article
A Noul returns one number and no separate confidence. How to read the value, pin its boundary with criteria, split compound questions, and set a per-action threshold that matches the cost of being wrong.
Jev Practices & Patterns#noul#primitives#confidence
Read article
A Score is a weighted position, not a ruler. Why the number can land between levels, why levels are judged separately, and how to combine several atomic scores in code instead of one crowded rubric.
Jev Practices & Patterns#score#primitives#scoring
Read article
From the Jev community
Posts from builders shipping with Jev right now.