Check architectural boundaries in agent-written code

Current registry release: v0.6.0. Verify the exact public identity with npm view [email protected] version dist.integrity before installation. The same exact version is active as the Lane-A trust anchor.

The release runs the engine's schemas on native Zod 4 (zod 4.6.7, the patched line) instead of the zod/v3 compatibility copy, and generates the published JSON Schemas with Zod's own z.toJSONSchema. Every committed artifact parses identically, and the behavioral changes are tightenings only: a decision timestamp must carry seconds and a Z or ±HH:MM offset, and numeric fields refuse unsafe integers and Infinity. It retains the stack foundation and selected declared-stack enforcement from v0.5.0. It does not add automatic repair or registry resolution.

Source candidate: v0.6.1 (not published). This compatible runtime patch updates the runtime dependency brace-expansion to 5.0.12 and adds no engine behavior. It is not on npm; install v0.6.0 above. The Lane-A trust anchor stays v0.6.0 until v0.6.1's own release and activation. The registry preflight below answers E404 until that release exists; do not install the candidate before it returns a version and integrity:

npm view [email protected] version dist.integrity

bce gives coding agents a local, deterministic architecture check. A versioned EngineeringBlueprint records the repository's structural rules. Agents draft those rules from your intent, inspect the code, and repair violations; the same engine checks every pull request. Humans own the intent and approve policy changes. Agents operate the day-to-day loop.

Released support (v0.6.0): TypeScript/JavaScript framework-surface AST extraction, direct TypeScript/JavaScript module boundaries, a Python import-surface MVP, and structured Python module boundaries. Node 22+ is required; the contract remains pre-1.0.

Evidence boundary: a score of 100 means the implemented checks passed over the observed surface. Route guard checks find governed call sites; they do not prove authorization, tenant isolation, or execution on every path. See the route evidence limits and correction, including the export forms missed by the immutable v0.3.0 release. Start with a bounded pilot and verify realistic violations in your own repository.

Watch BCE govern its own main branch: live GitHub stages, authenticated self-adoption, and explicit evidence boundaries.

Watch the context move

Explore the interactive C1–C4 spectrum · Watch the films · Read the static overview

Follow context narrowing into work, inherited rules surviving the zoom, and findings returning as evidence for review. This illustrates the wider reference architecture; it is not a claim that the complete platform loop is released in BCE. C1–C4 here means architectural zoom levels, distinct from the engine’s constraint type identifiers. The brief preview plays once; the full film and interactive experience have playback controls.

Start with your coding agent

Give your existing agent this task:

Read https://blueprint-conformance.github.io/bce/llms.txt and its agent start guide.
Inspect this repository and identify one supported architecture boundary from our stated intent.
Use BCE to draft the rule, prove a real RED → code fix → GREEN, and wire the local agent loop.
Keep setup advisory and present policy changes for my review. Report unsupported scope explicitly.

The agent start guide routes an existing gated repository straight to MCP run_gate {} and a new repository to local authoring and onboarding. Your current agent can do this with its own tools; BCE needs no additional model account for that path. The blueprint inspection tools and review cockpit let humans examine the same contract and evidence.

Run a real gate

Three commands. No account, hosted service, API key, or repository setup:

npm view [email protected] version dist.integrity
npm install --save-dev --save-exact [email protected]
npx --no-install bce demo

The released demo runs one conforming tree and one drifted tree. List six targeted recipes for extension registration, tenant access, egress, TypeScript and Python module layering, and configuration widening, then run the boundary closest to your repository:

npx --no-install bce demo --list
npx --no-install bce demo --recipe module-layering

Run the released First Win recipes, or keep the zero-argument proof above.

The architecture package

One checked-in blueprint defines the intended components, relationships, and boundaries. BCE extracts the repository it received, compares the two, and returns one deterministic verdict.

These are the specification's first four enforcing types: C1 requiredComponent requires a real pluginSurface; C2 requiredDependency requires its governed registration edge; C3 forbiddenDependency rejects the axios import; and C4 forbiddenPath keeps extracted components out of src/legacy/**. The taxonomy has four more enforcing types and three explicit reserved types—open the C1–C4 visual guide or read the exact semantics.

In v0.6.0, the AI-first review surface's bce propose writes an immutable draft packet to quarantine; the model cannot approve or land policy. Read the review ceremony.

Experimental compiler slice — source checkout

This source checkout routes forbiddenDependency through typed policy, fact, evaluation-plan and verdict IRs, with deterministic pass receipts and replayable proof links. Existing report formats remain unchanged; other constraint families retain their current evaluator.

Build this checkout with npm ci and npm run build to use the experimental API. It is not included in the released [email protected] npm package. Read the compiler readiness guide for the API, refusals, compatibility boundaries and deferred work.

See the gate discriminate

This excerpt is cut from a live engine run on every push. It keeps the decisive lines selectable while the full transcript retains every emitted detail.

$ bce gate --repo drift --blueprint-dir blueprint --extractor ast --all
::error::    - [no-direct-http-client/critical] extension:greeting.plugin
        observed: forbidden edge extension:greeting.plugin -> axios is present
        at:       src/greeting.plugin.ts#L16
bce gate [enforced]: 1/1 blueprint(s) evaluated, 1 failing.
$ echo $?
1

$ bce gate --repo clean --blueprint-dir blueprint --extractor ast
  ✓ [email protected] — score 100 (pass)
bce gate [enforced]: 1/1 blueprint(s) evaluated, 0 failing.
$ echo $?
0

CI derives those lines from real RED and GREEN runs and rejects byte drift. The proof contract also verifies the complete recording and visual replay against the engine.

One engine, three entry points

Use the CLI for local feedback, the pinned GitHub Action at the merge boundary, or ten read-only MCP tools inside an agent loop. They share the same extraction, evaluation, report, and exit-code path; policy changes remain outside MCP. The released Action source is pinned to blueprint-conformance/bce@0f33f4c00f7d4adc37cd4379fd7a6e86b65ec7de.

The Agent Skill loads these instructions on demand, prefers the read-only run_gate {} tool, fixes source code on RED, and re-runs the same contract. Inspect the skill, wire the done-check, or follow ordered onboarding.

Adopt without freezing the repository

Start in advisory mode, capture known debt in a shrink-only baseline, then enforce the same verdict. The mode is committed policy—not a skip flag—and moving backward requires a reviewed rationale. Read the brownfield adoption guide.

Evidence and limits

Mechanism evidence is strong; causal product benefit is not established. This repository has a generated suite whose current count is shown above, replayed RED/GREEN fixtures, 50/50 killed self-blueprint mutants, deterministic reports, and cross-platform CI. Those are first-party proofs on author-controlled infrastructure; independent witnesses remain 0.

The Stage D declared-stack replay record reports that an oracle-separated fresh-context agent run matched the recorded A–D matrix: pass, violation, structural refusal, and detached integrity verification. The raw evidence is not public, so this is an author-controlled record—not an independently verifiable community witness—and it does not change that zero count.

Accelerated pilot v6 retained all 16 paired attempts in one exact local model/client cell. Its author-operated record contains useful directional observations, including 2/8 versus 3/8 safe successful completions, but it is permanently ineligible for a product decision. One baseline infrastructure timeout remains in the intention-to-treat denominator. The Evidence Foundry v3 primary stage—120 paired tasks and 240 retained attempts—has not run; its later transport stages are separately gated. We do not claim that BCE makes agents more successful, cheaper, faster, or safer than a baseline. Inspect and replay the public evidence, check the public truth ledger, or inspect the study contract.

Start with your repository

The v0.6.0 release contains six packaged architecture recipes. Run one, then adapt it with a measured authoring walkthrough for an empty repository, plain JavaScript, TypeScript, a monorepo, or direct module layering: choose the boundary that must hold. The measured test keeps every layout's author → RED → fix → GREEN first win in under 60 seconds, including loaded-runner contention.

Specification: blueprint-conformance/v1alpha1 · Agent loop: MCP and agent workflow · Documentation: blueprint-conformance.github.io/bce

Released stack plane (v0.6.0). bce stack snapshot emits a content-addressed StackManifest of the declared dependency closure (npm lockfile v3/shrinkwrap and pnpm-lock v9, Dockerfile/compose image refs, the declared node runtime), and bce stack diff classifies every move between two manifests (added / removed / forward / backward / rewritten / flags-changed / spec-changed / unknown), failing closed on a backward, rewritten, or unproven move. A cross-OS golden-byte-identity proof and a built-dist-CLI RED/GREEN discriminating pair over a real revision pair now run in CI, and a read-only stack_diff MCP tool exposes the classifier over already-written manifests. Check the registry identity before choosing an install target:

npm view [email protected] version dist.integrity

The release also packages pnpm-lock v9 ingestion (sources[].parser widened accordingly) and five enforcing declared-stack constraint types: exact package pins, closure equality, forbidden packages, digest-pinned images, and allowed Node versions. Missing, opaque, variable, or conflicting required declarations refuse instead of passing. These stack-plane capabilities remain absent from immutable v0.3.1, while the enforcement types are absent from immutable v0.4.0.

Current registry release: v0.6.0. It runs the engine schemas on native Zod 4 and retains the stack plane above plus v0.5.0's expanded route-handler inventory, refusal of recognized unsupported or rebound handlers, and visible call-site evidence limit. A fresh registry installation passed all 52 route cases against the exact released archive, and a fresh public installation passed 90 declared-stack gate, CLI, MCP, replay, tamper, and refusal cases. This release also includes offline review preparation, explicit solo-steward lifecycle support, and the local specification; immutable v0.3.0 does not contain those additions.

Its exact npm integrity is sha512-iSQcNPM8RZuuqCK/M/ehnVQ9n18GNPI9YPjd9epNGMHgCjqGjn26bTw0lCreoLlOenEGTwH36DCpNwo6VCvfGA==, and its source/Action commit is 0f33f4c00f7d4adc37cd4379fd7a6e86b65ec7de. The canonical GitHub Release is immutable with the exact tarball, signed payload manifest, signed EvidenceRecord, and compliance report attached. Read the verification record. Compatibility remains pre-1.0.

Apache-2.0 — license, notice, and trademarks.