Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

jev-social

Local social research agent: Jev picks each bounded Instagram, TikTok or LinkedIn step and socai CLI runs it in your Chrome.

CLIbetaMIT
View on GitHub ↗Live demo ↗

Overview

Jev Social turns Instagram, TikTok and LinkedIn into source-cited research reports using your own signed-in Chrome. At each step Jev chooses one typed operation from the current allowlist: search, open a particular observed post or profile, read comments, download TikTok media when explicitly requested, or finish. The local socai CLI executes that operation in the browser, and the result is fed back so the next choice set is rebuilt from captured evidence.

It is an unofficial, local-first community project for people doing social media research who want each step grounded in observed posts rather than generated summaries. Code enforces confidence thresholds and rejects malformed or low-confidence decisions before execution. Reports cite captured posts, comments and video evidence, and the app streams post cards as it runs. It currently reaches Jev through OpenRouter by default; the README notes OpenAI Decisions API support is not yet wired.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich bounded socai operation and observed target should run next (search, open post/profile, read comments, download, or finish)?—

How it works

  1. Build an allowlist of operations and targets from captured browser state
  2. Jev selects one operation; low-confidence or malformed picks are rejected
  3. socai CLI executes it in the user's authenticated Chrome
  4. Captured evidence feeds the next choice until finish; report cites evidence

Strengths

  • Every step is a bounded, auditable operation
  • Reports cite real captured posts, comments and media
  • Local-first; runs in your own browser session
  • Has CI tests and versioned releases

Caveats

  • Unofficial project relying on your logged-in social accounts
  • Linux needs socai CLI built from source
  • Uses OpenRouter by default; requires a key

Getting started

With Node 22+ and Chrome signed in: `npx github:socai-io/jev-social#v0.1.13 onboard` then `npx github:socai-io/jev-social#v0.1.13`.

At a glance

Type
CLI
Status
beta
Language
TypeScript
Runtime
Node / local web app
Has code
yes
Has tests
yes
License
MIT
Stars
169

Use cases and integrations

How curated lists describe it

  • Social media research: uses a Jev `Choice` at each step to select a concrete socai CLI operation and observed post or profile target on Instagram, TikTok, or LinkedIn, rejecting malformed or low-confidence decisions before execution.

    yibie/awesome-jev

  • Browser-grounded social research: Jev selects bounded Instagram, TikTok, and LinkedIn search/read operations, socai executes them in the user's Chrome, and reports cite the captured posts, comments, and video evidence; unofficial community project

    AnotiaWang/awesome-decision-models

  • Jev-powered Instagram, TikTok, and LinkedIn research: typed routing, real browser evidence, streamed post cards, video capture, and cited socai reports.

    logicrw/awesome-jev-projects

  • Open-source, local-first social media research agent for Instagram, TikTok, and LinkedIn. Jev routes read-only steps; socai CLI captures cited browser evidence.

    wh000wh000/awesome-jev-live

  • Local Instagram and TikTok research app where Jev selects a platform and bounded socai operation from observed state, while code enforces confidence and the socai CLI executes the browser step.

    cobanov/awesome-jev

  • Local Instagram, TikTok, and LinkedIn research where Jev chooses bounded socai browser operations and streams captured evidence into cited reports.

    hellogumbo/awesome-jev

  • Instagram, TikTok and LinkedIn research where Jev chooses each next operation and a real Chrome session executes it, with the evidence kept.

    MrJev/awesome-jev

Frequently asked questions

What does Jev decide?

The next bounded operation, such as search, open a specific observed post, read comments or finish.

Does it post or interact on my behalf?

The operations described are read-only research steps plus explicitly requested TikTok downloads.

Which provider does it use?

TypeSafe Jev via OpenRouter by default; onboarding prompts for the key.

Related projects

  • jev-browser-use

    Local browser automation workbench with an Edge/Chrome sidebar and Codex MCP sharing one Jev-driven task engine, with user confirmation.

    MCP serverbetaTypeScript
  • jev-deepresearch

    Depth-crawling research agent where Jev makes every per-page relevance and next-link decision and an LLM only plans and writes.

    ProjectprototypeJavaScript
  • super-research

    Toy research agent where Jev makes every keep/skip and delve decision over search results, and one expensive call writes the report.

    ProjectprototypePython
  • Jev Deep Research

    Research agent where GPT plans and Jev finds evidence in parallel across document regions with Choice and Noul.

    Browser extensionresearchTypeScript
  • jcode

    RAM-efficient Rust coding-agent harness, listed as using Jev decisions for memory pruning, browser navigation and voice routing.

    ServiceproductionRust
  • public-browser

    MCP server and libraries that drive Chrome over CDP and hand a decision model like Jev a bounded menu of page actions.

    ProjectproductionTypeScript