Source: https://spec-kitty.ai/blog/spec-kitty-analyzer-v0-3-0-release

# Spec Kitty Releases spec-kitty-analyzer v0.3.0

> spec-kitty-analyzer v0.3.0 adds build provenance, anomaly triage, governance activity reporting, and release-note enforcement to improve visibility into AI-assisted development.

By [Emanuel Rose](https://spec-kitty.ai/about.md) · Published 2026-08-17 · Press release

This is a dated blog post. It may use retired names (Charter, Carta, doctrine) or describe features that are not in the product today; on any conflict, /llms.txt and the product pages win. Current product model: https://spec-kitty.ai/llms.txt

*Agent Analyzer (the `spec-kitty-analyzer` package) is a free companion tool, separate from the Spec Kitty CLI and Team Kitty. It reads your AI coding agents' session logs on your machine and reports where tokens and time go, with suggested fixes. The Spec Kitty CLI is the open-source tool that runs your agents through spec, plan, build and review.*

**Redding, CA – August 17, 2026** – Spec Kitty announced the release of spec-kitty-analyzer v0.3.0, an open-source governance analyzer designed to improve visibility into AI-assisted software development workflows.

The release adds automatic build provenance, a separate anomalies collection for unclassified signals, and an opt-in governance activity view that reconstructs spec-kitty-go verdicts from harness transcripts. These updates help teams trace reports back to the exact binary and commit, separate anomalies from confirmed failures, and review agent governance activity with more accuracy.

The new version introduces a structured build object across the version command and JSON reports, including `build.version`, `build.commit`, and `build.build_date`. It also adds anomaly triage for signals such as non-zero `exit_status` values and crash signatures, while keeping those signals separate from confirmed failure totals.

spec-kitty-analyzer v0.3.0 also includes the `go-activity` command and the `--include go` flag on `analyze`, allowing teams to review governance-hook verdicts such as ADMIT, DENY, DECISION_REQUIRED, and UNRESOLVED.

“AI coding agents are becoming part of real delivery work, but teams need a clearer record of what those agents did and how the work was governed,” said Robert Douglass, founder of Spec Kitty and contributor to the governance activity feature.

The release includes one documented JSON schema change: the top-level `version` field has been removed from `analyze`, `query`, and `missions` output. The analyzer version now lives at `build.version`, with `build.commit` and `build.build_date` available as additional fields.

Release notes are now enforced through `CHANGELOG.md`, which serves as the single source of record for release notes. spec-kitty-analyzer v0.3.0 is open source and available now on GitHub at [github.com/Priivacy-ai/spec-kitty-analyzer](https://github.com/Priivacy-ai/spec-kitty-analyzer).

## About Spec Kitty

Spec Kitty is a delivery control plane for AI-assisted software teams. It turns local agent work into visible, reviewable delivery progress, with specs, work packages, evidence, reviews, and launch readiness in one shared record. Learn more at spec-kitty.ai.

spec-kitty-analyzer is an open-source tool from the Spec Kitty team. It surfaces governance activity, compliance violations, and operational anomalies with full traceability to build artifacts.

Learn more at [spec-kitty.ai](https://spec-kitty.ai)

## Media contact

Emanuel Rose
emanuel@spec-kitty.ai
707 506 6138