jev
Skill4 starsChanged 52 days agoMITView on GitHub
Use when a coding agent needs a fast, $0 micro-decision (yes/no routing, confidence-gated execution), an adversarial plan review, a 4-channel research pass, lossless log compaction, or a repo/PR audit instead of ad-hoc free-text reasoning. Symptoms - "should I retry or escalate", "sanity-check this plan", "compact this giant log without losing paths/commands", "audit this repo before merge".
32 other projects use a version of this.
Adopted by
Most recently taken up by
- fralapo/awesome-agent-skillstheir own version, first seen Nov 7, 2026
- Lu1sDV/skillsmdtheir own version, first seen Nov 4, 2026
- josecabralf/skillstheir own version, first seen Nov 1, 2026
- dagonet/claude-code-toolkittheir own version, first seen Nov 1, 2026
- connorch/skillstheir own version, first seen Oct 28, 2026
- GastonGelhorn/jevmatetheir own version, first seen Oct 25, 2026
Used with
Projects with it also use review.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
SentrySkillreview
Fixture review skill.
Also in 5 projects that use this
Other versions32 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 32
- wuyoscar/jev-skill★ 665same name
- joetawil7/first-pass★ 128same name
- FrancoisChastel/jev-code★ 44same name
- BorisLeMeec/jev★ 38same name
- flc1125/skills★ 38same name
- pragma-sh/pragma★ 25same name
- ciberjohn/Hermes-Skills★ 17same name
- ratacat/ratacats-skills★ 9same name
- Vincent-the-gamer/skills★ 9same name
- AlexPEClub/Jev-Model-Router-Claude-Code★ 8same name
22 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repositoryOne other file it gives its agents.
- CLAUDE.mdCLAUDE.md
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 28 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 32
Run node dist/cli.js features for the full feature catalog and node dist/cli.js info to see which backend (local LLM / heuristic) is active.
- Very specific: line 39
If an MCP server is configured (see this plugin's mcpServers entry), prefer the equivalent jev_* tools over shelling out to the CLI.
“Common” is measured: kept by at least 6 in 10 of the 28 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Jev — offline decision layer
Wraps the Jev CLI (node dist/cli.js <command>, built from this repo) so routine
micro-decisions run through schema-validated primitives instead of free text.
When to reach for which command
| Need | Command |
| Yes/no or multi-option routing with a confidence score | decide "<question>" --options "a,b,c" |
| Sanity-check / stress-test a plan before executing it | redteam "<plan or assumption>" |
| Multi-step plan with hypothesis + anti-thesis scoring | plan "<goal>" --preset <preset> |
| Shrink a huge log/context without losing paths, commands, errors | compact --file <path> --goal "<what to keep>" |
| Web/academic/code/social research, Jev-ranked | research "<topic>" |
| Repo-wide architecture/security/legal/budget scan | audit <path> |
| Discover/rank Claude Skills for a task (installed + catalog) | skills "<task>" [--online] |
| Pre-execution guardrail for a risky tool call | guardrail --tool <tool> --args "<args>" |
| Run tsc+tests and record a reward for self-tuning | verify --cwd . |
| TRL/GRPO reward from helpfulness - toxicity | rljf --file <json with {prompts, completions}> |
| Guardrail an agent action against its allowed scope | scope-judge --intent "<intent>" --action "<action>" --rules "rule1|rule2" |
| Triage ad copy variants or a sales-call transcript | marketing --mode ad_copy --variants "v1|v2" / --mode sales_call --transcript "<chunk>" |
| Score us vs competitors across derived dimensions | competitor-matrix --file <json with {ourProductDescription, competitorTexts}> |
| Route a command-bar/voice/chatbot request local vs System-2 | jarvis-triage "<request>" [--state "<context>"] |
| Extract a PROJECT_PLAN.md checklist from an agent chat log | jarvis-auto-plan --file <log path> [--goal "<focus>"] [--writeTo PROJECT_PLAN.md] |
| Calibrated rating from an existing vision-LLM text description | media-judge --context "<what>" --description "<vision output>" |
Run node dist/cli.js features for the full feature catalog and
node dist/cli.js info to see which backend (local LLM / heuristic) is active.
Notes
- 100% offline-capable: no API key or GPU required, deterministic heuristic backend is always available as a fallback.
- If an MCP server is configured (see this plugin's
mcpServersentry), prefer the equivalentjev_*tools over shelling out to the CLI. - Full command/tool reference:
README.mdat the repo root.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.