Recently Evaluated

A macOS AI meeting notetaker that records, transcribes, and organizes conversations locally on-device, with summaries, threads, entity timelines, and semantic search.

A playful API that gives a coding agent a simulated “smoke break,” with endpoints for dispensing breaks and tracking the total count.

A Claude Code skill that lets the model inspect live camera frames so it can verify hardware displays, spot wiring issues, and check physical UI behavior on real devices.

Open-source causal-graph memory for AI agents with MCP support, scoring, active forgetting, and benchmark results published in the repo.

Shared AI-context platform for teams that captures decisions, gotchas, and domain knowledge once and makes it available to agents through MCP and popular coding/chat tools.

Telemetry is an observability platform for AI and LLM apps that tracks model calls, tool steps, retrievals, tokens, cost, latency, and errors. It supports OpenTelemetry ingest, TypeScript and Python SDKs, and integrations like Vercel AI SDK and LangChain.

Persistent cloud compute for AI agents, paired with an OpenAI-compatible model hub and usage caps from one account.

A Codex Skill that installs verified public-safe skin packs from GitHub releases, validates theme files, stages them locally, and shows rollback guidance before applying anything.

A Claude Code skill that launches a second model in a herdr split pane to adversarially review a diff, then checks each finding before returning a verdict.

Jaybase is an append-only fact store for AI agents handling critical business workflows. The repo explains its event model, retry and conflict handling, encryption, single-tenant design, and includes install and API usage examples.

A proof-of-concept governance harness for agentic coding loops, using machine-level policy enforcement, validation hooks, and a deterministic control point to keep agent outputs within approved rules.

A local proxy and CLI for switching and monitoring Codex and Claude Code accounts, with combined token-throughput analytics and rate-limit awareness.

TokenPath is an API for token-level citations and attribution for LLM outputs. It resolves claims in generated answers back to source spans, uses attention-based tracing, and includes an endpoint for raw heatmaps.

Local-first change control for agent skills, with evidence-gated promotion, SkillOpt optimization, and MCP-based serving of skills.

Browser-based infrastructure for running coding agents persistently, with workspace management, status tracking, and support for tools like Claude Code and Codex.

Headroom is an open-source context compression layer for AI agents. It compresses tool outputs, logs, files, RAG chunks, and conversation history before they reach the LLM, with support for a library, proxy, and MCP server.

Pull request adding TY25 support for moonshotai/kimi-k3 in TaxCalcBench through OpenRouter. The snapshot shows benchmark wiring, PDF handling details, saved outputs and evaluation reports, and regenerated charts/results for the full TY25 set.

Fractal is a developer tool for running hierarchical agent loops that break complex work into sub-tasks, spin up child nodes in git worktrees, and track state locally in SQLite with a terminal UI.

Self-hosted registry and analytics platform for internal AI components like Skills, MCP servers, agents, hooks, and prompts. The page explains the problem it solves, supported harnesses, and the basic server/CLI setup flow.

Google’s Gemini API guide for the latest models, with model comparisons, pricing, quickstarts, and migration notes for Gemini 3.6 Flash and 3.5 Flash-Lite.

A public research report from Inkling that maps Apple-to-OpenAI employee moves using profile data, with month-by-month trends, function breakdowns, title shifts, and examples of the roles that moved.

Lagotto Meter is a beta web tool that checks how an AI agent perceives a website versus what the site claims about itself. It positions itself as a semantic truth/readability analyzer with a reproducible analysis method and public tools.

Local, structured memory for AI coding assistants like Claude Code and Codex. The project stores decisions, constraints, and abandoned approaches in an event-sourced local log and injects them back into later sessions.

Open-source Python library for AI agent context governance with tamper-evident audit trails, provenance validation, freshness checks, and adapter support for common agent stacks.