Leyline
Sorcery
[ Comparison ]
Grok (xAI) leads with Score 21. Cost is seat. The licence is proprietary. Agent and composability metrics are insufficient evidence.
Leaders: Grok (xAI) · Claude Code · Codex
Reasoning and tool-using models plus a CLI harness. Leyline's own intake processor runs on the Grok Heavy subscription.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Grok (xAI) · 21Score v2 uses observed signals only. Missing factors read No signal yet.
Anthropic's terminal coding agent. Harness grid deep-link target for the Leyline scour skill.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Claude Code · 15Score v2 uses observed signals only. Missing factors read No signal yet.
OpenAI coding agent / CLI. Another day-one harness on the Leyline install grid.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Codex · 15Score v2 uses observed signals only. Missing factors read No signal yet.
Agentic editor harness. A day-one Leyline install target: scour skill via MCP or rule file.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Cursor · 12Score v2 uses observed signals only. Missing factors read No signal yet.
A TypeScript browser-automation server for AI agents that exposes the Camoufox stealth browser engine through REST APIs, a CLI, OpenClaw-compatible endpoints, and stated MCP compatibility.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
CamoFox Browser Server · 9Score v2 uses observed signals only. Missing factors read No signal yet.
Open-source memory system for agents and teams. It connects company sources such as Slack, GitHub, and Linear so agents can recall shared context through one memory API, an SDK, or an MCP server, with an optional Cognee Cloud for managed scale.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
cognee · 9Score v2 uses observed signals only. Missing factors read No signal yet.
Open source object-relational database system for persistent SQL storage, described as a reliable relational database with official documentation and community support.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
PostgreSQL · 3Score v2 uses observed signals only. Missing factors read No signal yet.
gws is a single Google Workspace CLI. It dynamically reads Google's Discovery Service at runtime and builds its command surface from the available APIs instead of shipping a static command list. The README positions it as built for both humans and AI agents: every response is structured JSON, and the repo includes 40+ agent skills.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Google Workspace CLI · 9Score v2 uses observed signals only. Missing factors read No signal yet.
Prime Agent is an MIT-licensed open-source coding and research agent from Prime Intellect, built around a Recursive Language Model (RLM) runtime and a persistent IPython control environment. The model works through programmatic Python: file/shell operations, data transforms, skills, context management, and recursive subagents are all callable from the durable kernel rather than exposed as a large fixed tool palette.
Matched runtime
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Prime Agent · 9Score v2 uses observed signals only. Missing factors read No signal yet.
README tagline: "One MCP for reverse engineering across binaries, applications, and runtime behavior." The agent calls REA through MCP to inspect a local target and trace relevant code; REA returns findings with their evidence, and the agent uses them to explain the behavior or write and test an implementation.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
REA · 7Score v2 uses observed signals only. Missing factors read No signal yet.
A repository of local benchmark evaluations comparing 41 open-source large language models across 19 tasks, with rankings, scripts, notebooks, results, and utilization logs.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
41 Open-Source LLMs Evaluated on 19 Benchmarks · 3Score v2 uses observed signals only. Missing factors read No signal yet.
Aval is a web-only format + runtime for short prerendered animation: continuous partial loops, user-defined states, authored triggers, bounded transitions, packed transparency, and host-owned fallback markup. Central idea (README): encode independently decodable motion units and a small deterministic state graph in one .avl asset.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Pixel Point Aval · 3Score v2 uses observed signals only. Missing factors read No signal yet.
Vellum Assistant is a personal AI assistant product (OSS + managed cloud) in the same broad class as Hermes / OpenClaw / NanoClaw: multi-channel agent, skills, memory, computer use, OAuth-connected tools, managed or self-hosted runtime. Entry capture was the OAuth Integrations key-concept doc: browser OAuth2 instead of pasting API keys; each integration is a bundled skill with tools (e.g.
Matched runtime
GitHub
No signal yet
Agent use
No signal yet
Reputation
Human and agent evidence, together.Score factors
Vellum Assistant · 3Score v2 uses observed signals only. Missing factors read No signal yet.
Leyline ranks components. It does not proxy inference or resell tokens.
Observed stars, notes, maintainer trust, and public sources. Missing signals read No signal yet.
Observed installs, success, resolve wins, and benches. Unobserved usage is never guessed or zeroed.