Skip to content
JevDirectory.org
CommunityTools & Integrations7 starsVerified 2026-09-22

pi-heed

A Pi extension that turns constraints stated in conversation into a scoped, replayable policy and checks every side-effecting tool call against it before it runs, with Jev only classifying how each message changes the policy.

Open github.comNyarlathoteppppp/pi-heed
Category
Tools & Integrations
Published by
Community
Author
Nyarlathoteppppp
Added
2026-09-22
Tagscommunitytypescriptguardrailscoding-agentagents

Highlights

  • Rules carry action, resource, scope, exceptions, and status as DENY, ALLOW, REQUIRE_CONFIRMATION, or REQUIRE_BEFORE, and the most specific resource wins.
  • On 79 scripted sessions and 261 labelled decisions, v0.8.0 with Jev reached 98.5% recall, 0.0% false block, 100% lifecycle, and 98.7% task success.
  • A replay of 21 real pi sessions with 9,090 tool calls blocked 95 calls across 16 incidents, down from 1,013 across 42 in v0.7.4.
  • Every policy change is an op persisted in the session, so state replays after reload or compaction without calling Jev again.
  • It fails open: a Jev error or 2.5 s timeout leaves pi behaving as if pi-heed were not installed.

Quickstart

bash
pi install git:github.com/Nyarlathoteppppp/pi-heed
npm install
npm test                          # 122 tests, no network
node bench/run.ts --judge replay  # benchmark, offline

Watch out

MIT-licensed; needs Pi and a TypeSafe API key or a local model, it is experimental and defaults to shadow mode, and the benchmark is scripted rather than drawn from real sessions.

More like this

pi-warden
132GitHub stars
Pi extension that supervises a coding agent with Jev judgments: it holds risky tool calls, checks writes against project Markdown rules, and feeds most issues back to the agent as a steer instead of interrupting you. The conscience is beta and off by default.
Tools & Integrations#community#typescript#guardrails
Communitypi-warden
Bicameral
5GitHub stars
A hybrid coding harness for Pi where any LLM writes the code while Jev returns typed probabilities that deterministic policy code turns into allow, confirm, block, warn, or steer.
Tools & Integrations#community#typescript#agents
Communitybicameral
jev-harness
1GitHub stars
A research-stage proposal-review contract in TypeScript: an LLM proposes one action, Jev answers four pinned questions, and code produces evidence the host can consider.
Tools & Integrations#community#typescript#agents
Communityjev-harness
Back to all resources

From the community

Posts from builders shipping with Jev right now.

Follow @typesafeai

A million judged questions

Inferring Jev's internals from 1,000 calls

iwashi / Yoshimasa Iwase
iwashi / Yoshimasa Iwase
@iwashi86
·

Jevの内部アーキテクチャを推測している技術記事(Jev’s Architecture Unmasked)からメモ。 ・本記事はJevのAPIを約1万回の呼び出して、内部構造を推測したもの ・従来の言語モデルを用いた分類やルーティングでは、トークンを1文字ずつ逐次生成するために膨大な無駄な計算コストが発生していた。 Show more

Reply
Read 2 replies

The open System One roundup

小墨同学
小墨同学
@xiaomovps
·

Jev 发布没几天，开源社区已经开始疯狂复刻了🔥 最值得推荐的五个模型： 1、Laya 421M：原生决策模型，支持 Mac 2、Decider-2B：最像 Jev，基于 Qwen3.5 3、NanoJev 0.6B：专门的 Decision Head 4、Reflex：Qwen3.5 + Direct Logits 5、System-One 4B：专门做概率校准 Show more

小墨同学
小墨同学
@xiaomovps

Jev 刚发布没几天，开源社区就出现了同款🔥 Decider-2B模型，是基于 Qwen3.5-2B 做了特殊调整 它和 Jev 模型是一样的 只做选择 评分和判断 不是文本类的 LLM 模型 但两者还是有几个明显区别： 1、模型 Jev：闭源 System One Model Decider：Qwen3.5-2B，约 1.9B 参数，Apache 2.0 开源 2、价格

Image
Reply
Read 6 replies

The launch post

Trading bot, one decision per block

Classifying 1,500 real emails

vogel
vogel
Anomaly
@ryanvogel
·

this model is actually insane at email classification i tested it on 1500 of my own emails to see how well it works and I am blown away

Watch on X
Diogo Almeida
Diogo Almeida
TypeSafe AI
@CompleteSkeptic

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x

Watch on X
Reply
Read 69 replies