Arcwell
local-first assistant infrastructure · private
Arcwell is the assistant infrastructure I wanted around Codex: durable memory, cited research, scheduled watchfulness and reusable skills, with its state kept together locally so I can see what it knows and what it is doing.
02codex-swift
native agent runtime
codex-swift treats Codex as a native Mac service, detached from any one terminal. It remains compatible with existing Codex clients while isolating sessions and leaning on Keychain, Seatbelt and launchd for the jobs macOS already knows how to do.
03open-cloud
local developer platform
Local tests should be able to exercise Cloudflare’s programming model entirely on the machine running them. open-cloud runs Workers and their bindings on a laptop or CI runner, keeping state in ordinary SQLite files and directories so failures can be inspected rather than chased through a dashboard.
04mem0-rs
agent memory substrate
mem0-rs moves mem0 2.0.4 from Python to a self-contained Rust binary because memory sits on an agent’s hot path. The port preserves the extraction prompts and retrieval behaviour that determine what the agent remembers.
05code-search
orientation before editing
Grep is excellent once an agent knows what a repository calls things. code-search gives it a reconnaissance pass first, using BM25-ranked results to teach it the local vocabulary before exact or regex search takes over.
06claude-code-goal
durable agent goals
It drives me nuts when an agent stops with half the job still sitting there. claude-code-goal keeps the objective in inspectable Markdown and uses Claude Code’s lifecycle hooks to continue the session until the work is complete or its iteration limit is reached.
07Interviewer
voice-first agent orchestration
A blank page is a poor way to discover what you know. I built Interviewer to draw it out in a voice conversation that feels like speaking with somebody who is paying attention; seven agents plan, listen, research, challenge and shape the exchange into a draft.
08ec-developer-dashboard
developer-program measurement
I built this to answer the question DevRel dashboards usually avoid: what did developers actually do after the event? It follows participants from their first contribution to deployed packages and mainnet transactions, so a programme can be judged by what it produced rather than how busy it looked.
Smaller tools
Narrower tools, each made to remove a recurring irritation from the way I build or test software.
-
security-scan-mcp
An MCP server that turns patterns from CVEs with known exploits into concrete review questions for the software an agent is inspecting. It replaces "look for security issues" with checks grounded in real vulnerabilities and exploit references.
-
devrel-almanac
A field guide for people starting, running or rebuilding Developer Relations programmes. It connects the field's history to decisions about organisation, measurement and tactics, including how the work changes when agents become part of the audience.
-
granite
I wanted the convenience of a modern knowledge app while keeping the vault under my control. Granite stores notes as ordinary UTF-8 Markdown, opens existing Obsidian-style content in place, and builds its navigation and editing around those files.
-
valkey-bun
valkey-bun gives local tests and small tools a Bun-native server that ordinary Redis clients can use over RESP2 or RESP3, removing the container that would otherwise exist merely to hold a few session hashes.
-
severe-testing
severe-testing asks a harder question than “does it work?”: what evidence would prove the claim false? It then sends agents after that evidence with independent oracles, hostile inputs, failure injection and recovery checks.
Standards and earlier specifications
Specifications and reference implementations for social applications, identity, authorisation, portable contacts, and API contracts.
-
Apache Shindig
I was the chief author of Apache Shindig's PHP implementation, a working gadget container, backend APIs and proxy for sites hosting OpenSocial applications. I also served on the PMC and as a committer.
-
OpenSocial Spec
OpenSocial defined a common API surface for web-based social applications across participating networks. I co-authored the specification so developers would not have to integrate separately with every platform's private API.
-
Portable Contacts
I co-authored Portable Contacts, a deliberately simple XML and JSON protocol for accessing address books and friends lists over HTTP, so developers could use one user-controlled pattern instead of integrating every provider's private API.
-
OAuth 2.0
I contributed to OAuth 2.0, the framework for granting applications limited, scoped access to an HTTP service on a user's behalf.
-
OpenID Connect
OpenID Connect adds an interoperable answer to "who signed in?" on top of OAuth 2.0 and makes basic profile claims available to the application. I contributed to its development.
-
OpenAPI
OpenAPI lets documentation, code generators and tests work from one machine-readable account of an HTTP service. I contributed to the API-description work that became the standard.