First-message LLM provider plugin for DeepSeek Harness session titles

Gate requests based on CORS data.

DSH permission-gate for the DeepSeek Harness 0.1.5 line: a single self-sufficient, deterministic-first, fail-closed gate covering P0 hard-deny -> P1 session grant -> P2 static rule (allow/deny) chain -> P3 optional LLM semantic classifier -> P4 ask, with

micromark extension to support math (`$C_L$`, `\(C_L\)`)

Ticket executability gate (P2) — a cheap-tier LLM checks whether a ticket is self-contained before a build agent starts.

[llm-ui](https://llm-ui.com) markdown block.

Hardware accelerated language model chats on browsers

LLM 게이트 저작 프레임워크 — 규칙을 넣으면 게이트가 된다

Shared LLM generation policy for DeepSeek Harness session-title providers

Model selection: the /model popupSelect over session.models / session.selectModel

Audits a spec for acceptance criteria that lack a concrete verification method (C1/P1).

Detox driver for Wix Pilot usage

[llm-ui](https://llm-ui.com) code block.

Lightweight LLM budget & token guard. Prevents Denial of Wallet attacks with a circuit-breaker state machine.

Much like tests in traditional software, evals are an important part of bringing LLM applications to production. The goal of this package is to help provide a starting point for you to write evals for your LLM applications, from which you can write more c

A work queue that can be gated, stopping to wait for sub-queues to complete.

Converts prosecution findings into permanent defenses — the compounding closer (P7).

Deterministic scheduler core for dossier batch cycles — queue, slots, persistent state machine, dispatch engine with completion verification and stall/escalation ladder, PR watching with script-based teardown, and batch failure recovery (attribution, bise

Process transparency for agentic work.

Google Gemini plugin for LiveKit Node Agents