Release 0.4.16 — RAG Truncation and Seed Persistence Fixes
Released: 2026-05-21 Chart:
oci://ghcr.io/cnoe-io/charts/ai-platform-engineering:0.4.16Previous release: 0.4.15
Highlights
0.4.16 is a maintenance release focused on correctness. The RAG_MAX_OUTPUT_CHARS cap now actually fires for MCP tool results (previously a type check silently skipped truncation), seed agent configs once again persist their ui, features, and interrupt_on fields to MongoDB, and two reopened CodeQL code-scanning alerts are closed. The bundled CAIPE API reference docs were also refreshed to match the current BFF and RAG surfaces.
What's New
Documentation
- Refreshed CAIPE API reference — replaced the stale
/api/usecasesand UI OpenAPI examples with a current BFF reference (auth, chat/runtime proxies, dynamic agent config, MCP server config, workflows, skills/catalog, RAG proxy, settings/users, admin routes) plus a dedicated Knowledge Bases RAG API reference (#1484)
Bug Fixes
- rag:
_truncate_outputnow serializes non-strresults withjson.dumpsbefore applying the length cap — MCP tools returnlist[dict]content blocks, so the previousisinstance(result, str)guard was alwaysFalseandRAG_MAX_OUTPUT_CHARStruncation never fired (#1488) - ui: persist
ui,features, andinterrupt_onfields inseedAgents— agent gradient themes, middleware config, and human-approval (interrupt) settings were silently dropped when upserting seed agents to MongoDB; the same change also refines the escalation-handoff and clear-role-definition micro-prompts and removes the no-second-person-preamble AI review rule (#1486)
Security
- code scanning: closed two CodeQL alerts that reopened on
main— removed a clear-text logging path derived fromclient_secretstate inget_token.py, stopped copying raw exception text into the skill scanner result dict inskill_scanner_runner.py, and restored the filesystem Grype SARIF category fromgrype-filesystemback to the historicalgrypecategory so older alerts are correctly superseded (#1478)
Breaking Changes
No breaking changes. Drop-in upgrade from 0.4.15.
Known Issues
None known at this time.
Upgrade
helm upgrade ai-platform-engineering \
oci://ghcr.io/cnoe-io/charts/ai-platform-engineering \
--version 0.4.16 \
-f your-values.yaml
Upgrade Guide: 0.4.15 → 0.4.16
Overview
Drop-in upgrade — no values.yaml edits required. 0.4.16 ships only application bug fixes (RAG output truncation, seed-agent field persistence) and a security fix for code-scanning alerts. There are no chart, schema, or configuration changes.
Helm Values Changes
No Helm values changes between 0.4.15 and 0.4.16. Drop-in upgrade — no values.yaml edits required.
Data Migrations
No MongoDB schema or data migrations required.
Upgrade Runbook
1. Update chart version
helm upgrade ai-platform-engineering \
oci://ghcr.io/cnoe-io/charts/ai-platform-engineering \
--version 0.4.16 \
-f your-values.yaml
2. Verify
kubectl get pods -n <namespace>
All pods should roll over with new image tags and return to Running. No further action is required.
