feat(runtime): distribute canonical resumable capability kernel

Merged
GL
2mo ago
+5284-543

User description

Runtime completion

Distributes GlacierEQ/apex-boot-core@84a2907a316327e91dc0426f5407a34908aa4fc4 as the sole canonical runtime authority.

  • Transparent pinned-source projection: 12 files, every SHA-256 verified at install.
  • Composition receipt separated from canonical runtime receipt.
  • apex boot runs capability probe → lifecycle checkpoints → canonical receipt → replay attestation.
  • apex replay, apex resume, apex capabilities, apex runtime-health, and apex runtimectl are first-class.
  • Public Truth now proves fresh install, projection hashes, canonical commit attestation, TTL invalidation, inventory truth, resume cursor, idempotent side effects, resource governor denials, and runtime health.

Apex Boot Core canonical local suite: 14/14 PASS; canonical projection verifier: 12/12 PASS.

Summary by Sourcery

Distribute Apex Boot Core as the canonical runtime and separate composition from runtime lifecycle, receipts, and attestation.

New Features:

  • Introduce transparent canonical runtime projection download pinned to a specific GlacierEQ/apex-boot-core commit with SHA-256 verification at install time.
  • Add a dedicated runtime CLI pipeline from startup for capability registration, probing, replay, metrics, and canonical receipt emission.
  • Define composition receipts as a distinct artifact from runtime receipts via a portable composition adapter and updated schemas.
  • Expose new runtime control and health capabilities including capability inventory management, resumability, idempotent action tracking, governor enforcement, and runtime health reporting.

Enhancements:

  • Tighten installer verification to assert presence and correctness of canonical runtime projection, scripts, and contracts, and to validate the runtime via apex boot, runtime-status, and replay.
  • Refine startup and composition error handling, exit codes, and logging to clearly distinguish composition adapter failures from runtime failures.
  • Record composition latency metrics and persist boot-time skill and predicate state into the runtime state directory for future attestations.

Tests:

  • Extend Public Truth CI workflow to prove canonical runtime installation, projection verification, capability freshness and inventory truth, resumability and idempotent side effects, governor economics, and runtime health semantics.
  • Update CI shell boundary checks to assert the presence of new runtime subcommands like apex replay and apex resume.

CodeAnt-AI Description

Make the canonical resumable runtime authoritative for boot, receipts, and recovery

What Changed

  • Installs the pinned Apex Boot Core projection and verifies every downloaded file against its expected checksum.
  • Boots through canonical lifecycle checkpoints, capability checks, runtime health tracking, resource limits, and final runtime receipts.
  • Adds commands for replaying receipts, resuming incomplete runs, inspecting capabilities, checking runtime health, and viewing governor state.
  • Separates skill composition evidence from canonical runtime receipts, with clearer composition-specific errors and status output.
  • Records boot skills, manifest checksums, composition latency, resumable cursors, and idempotent action results for later verification.
  • Expands public proof coverage for installation integrity, capability freshness, resume behavior, duplicate side effects, resource denials, and runtime health.

Impact

✅ Verified canonical runtime installs
✅ Resumable boot after interrupted lifecycle stages
✅ Clearer runtime and composition status

💡 Usage Guide

Checking Your Pull Request

Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.

Talking to CodeAnt AI

Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:

@codeant-ai ask: Your question here

This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.

Example

@codeant-ai ask: Can you suggest a safer alternative to storing this secret?

Preserve Org Learnings with CodeAnt

You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:

@codeant-ai: Your feedback here

This helps CodeAnt AI learn and adapt to your team's coding style and standards.

Example

@codeant-ai: Do not flag unused imports.

Retrigger review

Ask CodeAnt AI to review the PR again, by typing:

@codeant-ai: review

Check Your Repository Health

To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.

Summary by CodeRabbit

  • New Features
    • Added runtime controls for status, health, capabilities, replay, resume, and resource governance.
    • Added verified canonical runtime installation with integrity and commit checks.
    • Startup now tracks runs, checkpoints, receipts, resumable actions, and runtime health.
  • Bug Fixes
    • Improved fail-closed handling, validation, and runtime error reporting.
  • Tests
    • Expanded coverage for replay, quotas, eligibility, schemas, health, idempotency, and commit identity.
    • Added automated runtime proof, installer synchronization, and validation workflows.
    • Increased verification timeout for improved reliability.

Summary by cubic

Distributes a SHA-verified, pinned canonical runtime projection and separates composition receipts from runtime receipts. Previously runtime and composition shared a mutable projection; now a one-to-one projection installs atomically, emits distinct receipts, supports replay/resume/health, and merges only with a VERIFIED public proof. Adds tests proving invalid JSON inputs fail closed.

  • Review and rollout

    • Enforces synchronized public installers: install.sh equals install-runtime-v2.sh. CI adds Sync Runtime Installer, a clean-home Final Runtime Proof, and extends Public Truth to master and the integration branch; merges gate on RUNTIME_PUBLIC_PROOF.json; post-merge cleanup removes temporary transport scaffolding.
    • Ships a public runtime adapter and projection verifier; uses pinned raw-source transport with SHA-256 digests, duplicate-path rejection, unique-hash selection, atomic projection replacement, recovery from the pinned source, and preserved capability-inventory digest. Exposes runtimectl, replay, resume, capabilities, runtime-health, governor, and runtime-status. Normalizes invalid JSON file inputs in adapters to safe defaults and proves fail-closed behavior in tests.

  • Migration

    • Read runtime receipts from ~/.apex/last_runtime_receipt.json and composition receipts from ~/.apex/last_composition_receipt.json.
    • Update callers of runtime-status to expect canonical runtime identity fields only.
    • For offline installs, pre-seed ~/.apex/runtime_projection with the canonical package; non one-to-one or mismatched transports are rejected.