This release fixes issues for SREs watching stability and regressions.
Published 1mo
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 summaryFable model pins now resolve to Opus consistently, fixing cost mismatch across dispatch, quoting, and UI.
Full changelog
Fixed
- Fable model (claude-fable-5) is not available at the Claude API ("Claude
Fable 5 is not available, use Opus 4.8"). A session pin, mid-flight override,
or architect opt-in to fable now resolves to opus consistently across all
three readers: the dispatched model (providers/claude.sh resolve_model_for_tier,
autonomy/run.sh static fallback and the final pre-argv backstop,
loki-ts claudeTierToModel and rarv.ts getProviderTierParam), theloki plan
cost quote, and the dashboard effective-model. A fable pin now both runs AND
quotes opus ($5/$25), not fable ($10/$50), so the quote matches the dispatch
(cost honesty). The fable tier label, session-pin parsing (fable stays a valid
input alias), and pricing-table rows are preserved; only the resolved model
changes. The 224-cell session-pin parity matrix stays green (dispatch == quote
== dashboard for every fable cell).
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]