This release adds 2 notable features for engineering teams evaluating rollout.
Published 26d
AI Agents & Assistants
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
ai-agents
aider
anthropic
autonomous
ci-cd
claude
+13 more
cline
code-review
devops
gemini
github-action
github-issues
loki-mode
multi-agent
openai-codex
openapi
pull-request-review
sdlc
spec-driven-development
Summary
AI summaryBun runner now blocks completion on code_review, semantic_tests, and invariants gates matching bash behavior.
Full changelog
Bun-runner gate parity + verify.sh hardening
- Bun runner honors all completion-blocking gates (
loki-ts/src/runner/autonomous.ts):
it already refused completion on a code_review BLOCK; it now also refuses on a
semantic_tests BLOCK and an invariants BLOCK, mirroring the live bash route
exactly (sameLOKI_GATE_SEMANTIC_TESTS_BLOCK/LOKI_GATE_INVARIANTS_BLOCK
toggles, same conditions). This closes a trust-parity gap ahead of the runner
being wired on. It refuses ONLY when the gate genuinely blocked (the gate is in
the failed set and its toggle is on), so a clean run still completes - no new
user-visible knob, the existing toggles are reused. - verify.sh defensive hardening (
autonomy/verify.sh): the base ref now
resolves an actually-existing ref (origin/main, then main) instead of
hardcoding it in both branches; VERIFY_VERDICT/VERIFY_EXIT are initialized
before the error path so an early error can never emit an empty or VERIFIED
verdict or exit 0 (fail-closed).
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Loki Mode
Multi-agent autonomous SDLC framework. Spec to deployed app. PRD, GitHub issue, OpenAPI/JSON/YAML, or one-line brief. 5 AI providers, 11 quality gates.
Related context
Related tools
Beta — feedback welcome: [email protected]