Scaffolde: A personal AI capability factory

The thesis

Scaffolde is the factory. Runtime surfaces are projections.

Every new agent runtime arrives with its own config directory and its own copy of your capabilities. The durable asset is not the runtime, it is the versioned capability catalog upstream of all of them.

A factory, not a chat client

Capabilities authored as versioned source outlive any single model, vendor, or CLI. Scaffolde sits upstream of them all: skills, agents, workflows, runtime services, and verification machinery in one canonical repository.

Write once, project everywhere

Typed manifests plus a one-way sync engine put one capability catalog onto 12 declared surfaces. Drift detection keeps the repository authoritative: a hand edit under a runtime root is drift, not authority.

Discipline as moat

Evidence-backed completion is the practice that keeps a compounding system trustworthy to its operator: falsifiable claims written before the work, fail-closed gates, and 10,900+ maintainer-health tests green.

Market & why now

Every new agent runtime is another config directory.

Agent tooling multiplied faster than anyone built a place to keep capability. The result is the same skill, rewritten per tool, drifting per machine.

Runtimes multiplied

Claude Code, Codex, Cursor, and a wave of agent CLIs each keep their own skills, prompts, and settings under their own root.

Capability got fragmented

The same skill gets rewritten per tool, drifts per machine, and dies with the session that produced it.

Hand edits don’t scale

Live tweaks under a runtime root such as ~/.claude are unversioned, unreviewed, and silently lost on the next update.

The missing piece is upstream

A canonical factory that versions capability once and projects it everywhere, with drift detection and evidence deciding what returns to source.

Projection pipeline

1Canonical source
skills, agents, workflows, adapters, versioned in one repository
2Typed manifests
declare what each surface receives; missing references stop projection
3Declared surfaces
one-way sync writes managed output, with drift detection before any overwrite

Claude CodeCodexHermesOpenClawPrime AgentPaperclip+ 6 more declared

Fail-closedDrift-detectedDeclared, not a health guarantee

The wedge

One operator. One capability source.

Beachhead: the individual AI-native operator running multiple agent runtimes who needs one versioned capability source instead of five drifting config directories.

The founder is user zero: the system runs his work daily, and every capability on it was authored, projected, and graded by the same loop. Productization expands from one operator to teams, where the same catalog and the same evidence rules apply to shared work.

See what the factory owns →

Why this founder

Built by someone who has shipped enterprise systems under real constraints.

Gary Noonan

Founder · Sr. Principal Enterprise Architect, Comcast Advertising

Gary has spent his career architecting large-scale enterprise systems where engineering decisions have to hold up to scrutiny. He built Scaffolde as the factory behind his own AI work, and is raising to make it a product.

Commitment: Gary is transitioning to Scaffolde full-time on close of the pre-seed round. This is the company, not a side project.

Traction

A working system, honestly measured.

Counts come from the repository’s generated indexes and manifests, August 2026. Status is stated plainly: private, single-operator, macOS, v2.0.0 release candidate.

Working system

Private

Scaffolde runs the founder’s daily AI work today: 177 active skills, 51 scheduled jobs, and 12 declared surfaces, maintained as a private single-operator deployment on macOS.

Running daily

Engineering evidence

Green

10,900+ maintainer-health tests passing across 991 files, 34 gate validators in a registry CI-asserted against the validator file set, and a release gate that fails closed. v1.0 and v1.1 tagged; v2.0.0 is a release candidate.