ingero-io/ingero

v0.19.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 10d MCP Data & Storage

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

causal-tracing cuda cuda-graphs ebpf gpu gpu-monitoring

+11 more

gpu-observability incident-response kubernetes machine-learning mcp model-context-protocol nvidia observability pytorch sre distributed-tracing

ReleasePort's take

Light signal

editorial:auto 10d

ReleasePort v0.19.0 introduces per‑request annotation keys and new CLI flags (explain --by-request, query --by-request).

Why it matters: These additions enable more granular request tracing across the CLI, contract schema, events storage, store API, MCP, and dashboard, improving observability for developers, SREs, and security engineers.

Summary

AI summary

Adds per-request annotation keys, explain --by-request flag with honesty caveats, and query --by-request flag.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Adds explain --by-request flag with honesty caveats to CLI. Adds explain --by-request flag with honesty caveats to CLI. Source: llm_adapter@2026-05-24 Confidence: high	—
Feature	Medium	Adds query --by-request flag to CLI. Adds query --by-request flag to CLI. Source: llm_adapter@2026-05-24 Confidence: high	—
Feature	Medium	Adds v0.19 per‑request annotation keys to contract. Adds v0.19 per‑request annotation keys to contract. Source: llm_adapter@2026-05-24 Confidence: high	—
Feature	Medium	Persists NCCL barrier_wait event to the events table in CLI and emitter. Persists NCCL barrier_wait event to the events table in CLI and emitter. Source: llm_adapter@2026-05-24 Confidence: high	—
Feature	Medium	Adds SelectEventsByLabelValue query for windowed by‑value grouping in store, MCP, and dashboard. Adds SelectEventsByLabelValue query for windowed by‑value grouping in store, MCP, and dashboard. Source: llm_adapter@2026-05-24 Confidence: high	—
Feature	Medium	Adds per‑request summary entries in commands, detections, and vLLM README documentation. Adds per‑request summary entries in commands, detections, and vLLM README documentation. Source: llm_adapter@2026-05-24 Confidence: low	—
Bugfix	Medium	Quotes label values outside the safe charset in CLI formatLabels function. Quotes label values outside the safe charset in CLI formatLabels function. Source: llm_adapter@2026-05-24 Confidence: low	—

Full changelog

Changelog

Documentation

c7bd321878cc110387a21db7d035ffcf2a2d7e21 docs: per-request summary entries in commands + detections + vLLM README

Other

6c5974d9a70b4a6c5e9a4da02599971261972bef cli + emitter: by-request flag-combo guards and emitter scope defaults
7c27f8d130839438ce1335949e04b69c41304174 cli: Go-quote label values outside the safe charset in formatLabels
8e998f80c28059cd985afcb98035ebdef010c55d cli: add explain --by-request flag with honesty caveats
5955d9ed1fd68baa0e08e775313e2a679a5686fe cli: add query --by-request flag
4e5428c7de104a83a5e0d44d6049e109f9351987 contract: add v0.19 per-request annotation keys
81bd38d62a6380355e286ea794324e4842d7f2f6 events + cli: persist NCCL barrier_wait to the events table
1b18bba169efa7f9baecc8d9d2287ae47004c008 examples/vllm: per-request annotation emitter
b69cbcba0005aa563a5449273459d8c09bc9e98a python: ingero-annotate end-to-end install-and-use tests
467e0586ae21253982c70dc4804c368d930ea9fc python: ingero-annotate package; consolidate vendored adapter copies
0afc304d4b63ec3827ecdba2c231451dc2ffc3fa store + mcp + dashboard: NCCL schema seed and ecosystem rendering
c2b269b1d74d7524eed8d85a387eefff1137af5e store: add SelectEventsByLabelValue for windowed by-value grouping

View diff on GitHub

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

Share on X Share on Bluesky

Track ingero-io/ingero

Get notified when new releases ship.

About ingero-io/ingero

eBPF-based GPU causal observability agent with MCP server. Traces CUDA Runtime/Driver APIs and host kernel events to build causal chains explaining GPU latency.