For regulated teams · CTO / CPO

Everything GraphSmith does — and the paper trail to prove it.

The full capability set, the five protocol invariants, and the complete 11-Document Build System — for teams shipping agents into finance, health, identity, and other regulated work. Local-by-default. No telemetry. Evidence on every run.

Local by defaultNo telemetry, everHuman gates on irreversible actionsTamper-evident evidenceNo lock-in
The protocol

Five invariants, enforced at every boundary.

v0.2.0 turns GraphSmith's discipline into a protocol. A deterministic check runs at every boundary — a workflow either passes graphsmith verify or it doesn't.

01
Verified integrity

A deterministic check runs at every boundary; injected text in a log or tool output can't reach control flow.

02
Immutable, gated core

A constitutional core no automated process can edit; every change passes a four-gate pipeline before it's real.

03
Isolated evaluation

Every candidate change is tried first in a disposable, mocked-effect copy; nothing touches your live project until it's earned it.

04
Local by default

Healing, learning, and evolution stay on your machine. Nothing is sent anywhere. No telemetry, ever.

05
Observable & killable

Every run declares budgets, streams to a live view, and is safe to kill and resume; tripwires halt a run heading somewhere dangerous.

The full inventory

Everything in the box — nothing behind a tab.

One install. Every capability across the lifecycle, laid out in full for evaluation. All local, all dependency-free, all readable in minutes.

Build
Scaffolder

A runnable, zero-dependency project: deterministic manager, worker steps, fsync-durable save points, resume, capped retries, structured logs — runs immediately with no API keys.

Blueprint gate

A one-screen plan — workers, handoffs, save points, stop rules — that you approve before a single line is written.

Per-run locking

A lease + heartbeat so a second manager on the same run refuses while the first is alive; a dead holder's lock is stolen automatically, so a crashed run self-recovers.

Fix
Architecture linter

Import-graph-aware scan of existing JS / TS / Python for the classic failure patterns — unbounded LLM loops, missing persistence, clock/randomness near control flow, unkeyed external writes — ranked by severity.

Self-tested corpus

--selftest runs the linter against a bundled corpus (every published adversarial-review probe is a case) and fails on any precision/recall regression.

Smallest-fix diagnosis

Maps "it forgets / duplicates / loops forever" to the exact broken rule, with file and line — and proposes the minimal fix without rewriting what works.

Verify
Chaos harness

Kill test, double-run, power-loss probe, three lock probes (dead-holder, live-recycled-pid, lease-expiry), and a loud safety-halt path. A forged halt fails — the verdict comes from on-disk state, never a string.

graphsmith verify

Reports capability profiles — resumable, effect-reconciled, budget-enforced, integrity-verified, gated-learning, assurance-tested, adversarially-tested — with linked evidence.

CI enforcement

Enforceable on every pull request through a GitHub Action or a GitLab template. A workflow passes the gate or it doesn't merge.

Ground
KnoSky citations

A local pointer index your AI cites from — claims about your code carry a real file reference, or are flagged as guesses. A map, not an oracle. Nothing leaves your machine.

Supply-chain hardening

KnoSky is pinned by version AND content hash (refuses on registry mismatch), never a silent global install. GRAPHSMITH_OFFLINE=1 skips all network.

Read-only by design

KnoSky indexes pointers (titles, headings, short excerpts), never file bodies. Its index writes to a local .knosky/ folder in your repo.

Coordinate
Lanes

Each agent owns a module boundary; one writer per lane, ever. Merge conflicts are planning failures to fix in the plan, not battles to win in the diff.

Claims with leases

Work lives in a shared task file; an agent locks a task atomically, and a dead agent's expired lease frees it. No task is worked twice or orphaned silently.

Frozen contracts & no self-cert

Interfaces freeze before dependent work starts; the agent that built a thing never reviews it — the checker runs on a different model family.

Harden · v0.2.0
Self-healing

A halted or broken workflow gets a diagnosis and a smallest-fix repair, staged with evidence and one-command rollback. Code repairs stage for a human — never a silent rewrite of your logic.

Gated learning

Bounded, project-local improvements that must clear an executable gate and a human before adoption. It cannot learn its way past its own safety rules.

Assurance testing

Architectural, unit, smoke, regression & adversarial batteries against your own workflows, with tamper-evident evidence. A passing battery is a floor, not a certification.

The framework for production teams

The 11-Document Build System.

For teams running many agents against production software — real users, money, or regulated data — GraphSmith ships the complete agent-first delivery pipeline: a document for every stage from PRD to evals, each machine-readable and traceable by ID. Adopt it when your project earns it, not before.

#
Document
Purpose
00
Repo Map & Document Registry
Where every artefact lives; how agents find and trust documents (manifest + KnoSky index).
01
PRD
Problem, users, scope, requirements, success criteria — and the criticality that decides later whether a change auto-deploys or waits for a human.
02
Technical Design
Architecture, stack, module boundaries, contracts, constraints and trade-offs.
03
App Flow & State Map
Screens, journeys, states, actions and redirects.
04
UI/UX Brief
Visual system, interaction rules and accessibility.
05
Backend Design & Data Model
Data model, access rules, API contracts, storage and events.
06
Engineering Plan
Task graph with lanes and dependencies, built for parallel agents.
07
QA & Adversarial Test Plan
Full test taxonomy, adversarial charter, synthetic data, quality gates.
08
Release, Deploy & Operations
Environments, pipeline, progressive delivery, rollback, observability, runbook.
09
Agent Operating Protocol
Roles, pipeline state machine, risk tiers, human-in-the-loop gates, escalation and audit.
10
Evals & Model Performance
Scorecards for every stage and agent, hallucination controls, eval-gated tuning.

The seven rules it runs on

Fast path. A weekend prototype may combine everything into one lean file. A production app handling payments, personal data, or several roles needs the full set — and the manifest (00) and risk tiers (09) are the last things to drop: they're what keep parallel agents and automatic deploys safe.
Trust & assurance

Conservative by design — and honest about its limits.

The differentiator isn't a bigger claim; it's a smaller, provable one. Honesty is the enterprise sell.

No lock-in

A documented off-ramp, not a walled garden.

GraphSmith is the disciplined on-ramp, not a competitor to durable-execution engines. The graduation ladder tells you exactly when a workload outgrows this tier — and the migration seam is already isolated in the generated code.

Rung 1
JSON-file checkpoints

Correct for single-machine jobs up to roughly thousands of steps. Where every new project starts.

Rung 2
SQLite

When step counts and log volume grow past the file tier. Same conventions, more durable store.

Rung 3
Framework checkpointer

A LangGraph-class checkpointer when you need its ecosystem and state model.

Rung 4
Durable execution engine

Temporal / Inngest for distributed, long-running, exactly-once workloads at scale.

Evaluate it

Install it. Try to break it. Read the reports.

Free, MIT, zero-dependency, no API keys. Time-to-first-value is under a minute.

npx skills add SathiaAI/graphsmith