Skip to content

Releases

CodeStrata Community Edition v0.2.1

Choose a tag to compare

@sknampally sknampally released this 14 Aug 11:55

CodeStrata Community Edition v0.2.1

Patch release of the Community assessment CLI.

Highlights

  • Consent v2 shared foundation for CLI and VS Code consent surfaces
  • Privacy-safe assessment intelligence (no source upload / no repo identity in cloud payloads)
  • assessment_metadata 1.1 client support for richer voluntary assessment intelligence
  • Generated report favicon packaging
  • Transparency and documentation improvements for Community Edition use

Install

pip install codestrata==0.2.1
codestrata --version

Notes

  • This GitHub release attaches the exact frozen wheel/sdist used for PyPI publication.
  • VS Code Marketplace extension updates are published separately and are not part of this Engine/CLI artifact release.
  • Backend/Insights operational performance work is independent of this Python package identity.

CodeStrata Community Edition v0.2.0

Choose a tag to compare

@sknampally sknampally released this 11 Aug 04:19

CodeStrata Community Edition v0.2.0

Community public source: CodeStrata Engine and CodeStrata Examples.
CLI is distributed as a package. The VS Code extension is Marketplace distribution (source private). Platform and infrastructure remain private.

Added

  • Deterministic multi-head Engineering Assessment with Assessment Overview
  • Community report publishing with opaque public report URLs
  • Privacy-first telemetry consent (default deny; opt-in is not publish authorization)
  • Voluntary report feedback / Community Sentiment
  • Community Cloud and Insights
  • VS Code extension distribution via Marketplace
  • AI provider platform: AWS Bedrock live E2E validated; OpenAI/OpenRouter supported with documented owner-E2E qualification
  • Transparency and privacy documentation
  • Final 22-repository production validation corpus

Notes

  • Deterministic findings remain the source of truth. Optional AI enrichment does not invent findings.
  • OpenAI/OpenRouter owner live E2E was not completed; those providers remain supported with qualification.
  • This release does not publish Platform, Infrastructure, or VS Code source.

CodeStrata Community Edition v0.1.0

Choose a tag to compare

@sknampally sknampally released this 29 Jul 17:27

CodeStrata Community Edition v0.1.0

First public release of CodeStrata Community Edition — an open-source CLI that assesses a software repository and produces deterministic findings, recommendations, and self-contained Engineering Assessment reports in HTML and JSON, together with findings artifacts.

The Engine produces structured engineering intelligence.

Community Edition scope

Community Edition supports local, single-repository assessment:

  • Technology detection and evidence extraction
  • Deterministic rules, findings, and recommendations
  • Self-contained HTML and JSON reports
  • Optional MCP server support
  • Optional AI narrative through --with-ai

AI is not required for the basic Community workflow.

Community Edition does not include hosted SaaS or CodeStrata Platform capabilities such as Engineering Knowledge Graph, Repository Retrieval and Answering, Portfolio Intelligence, Executive Intelligence, or Strategic Roadmap.

Public repositories

Repository Role
codestrata-engine Assessment CLI and Engine documentation
codestrata-examples Pinned real-world showcase manifests and fetch/showcase scripts

Requires Python 3.12 or newer.

Install

git clone https://github.com/CodeStrata/codestrata-engine.git
cd codestrata-engine

python3.12 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e .

codestrata version
codestrata init
codestrata doctor

If python -m pip fails after activation (for example a shell python alias), use:

.venv/bin/python -m pip install --upgrade pip
.venv/bin/python -m pip install -e .

Offline sample assessment

codestrata assess --repo test-fixtures/sample-js-app --output reports --no-ai

Generated under reports/sample-js-app/<timestamp>/:

  • report.html
  • report.json
  • findings.json (and related assessment artifacts)

Basic analysis is deterministic. Use --no-ai for zero provider calls — AI is not required.

Optional MCP

python -m pip install -e '.[mcp]'

MCP is optional and not required for Quick Start.

Standalone PetClinic showcase

git clone https://github.com/CodeStrata/codestrata-examples.git
cd codestrata-examples

python real-world/scripts/run_showcase.py spring-petclinic

Or fetch then assess:

python real-world/scripts/fetch_example.py spring-petclinic
codestrata assess \
  --repo .codestrata-examples/spring-petclinic \
  --output reports/showcases/spring-petclinic \
  --profile community \
  --no-ai

codestrata-examples ships a root codestrata.toml so this works from a standalone clone. Showcase runs need network access to fetch the pinned upstream commit.

Known limitations

  • Bundled offline fixture: sample-js-app only in this release
  • Real-world showcases require network access to pinned third-party repositories
  • Optional AI narrative needs separate provider credentials and is outside the basic verified workflow
  • Language samples beyond the published JS fixture are not part of this Community release surface

Contributing and feedback

  • Open issues and pull requests on codestrata-engine for the CLI and Engine docs
  • Open issues and pull requests on codestrata-examples for showcase manifests and scripts
  • Security reports: see SECURITY.md in this repository
  • License: MIT (LICENSE)

Thank you for trying CodeStrata Community Edition.