claude-grc-engineering

https://github.com/user-attachments/assets/a83aa297-9fba-4a7d-b56c-06f962d1ec6b

Open-source GRC Engineering resource for Claude.

claude-grc-engineering turns technical evidence from cloud, SaaS, code, and security tools into framework-aligned findings, gap reports, remediation guidance, evidence packages, and OSCAL workflows.

It is built for the Claude ecosystem: Claude Code plugin installs first, with Claude Desktop and Claude Cowork usage supported through the same Markdown skills, command runbooks, schemas, and repository files.

It is maintained by the GRC Engineering Club for people who want compliance work to behave more like engineering work: repeatable, testable, versioned, and easy to extend.

Not affiliated with Anthropic. Claude, Anthropic, and related marks are property of their respective owners.

What it does

The toolkit is a Claude Code plugin marketplace. The same plugin skills and command runbooks are also useful in Claude Desktop and Claude Cowork when you add this repository as project context or a shared workspace. Install the pieces you need:

• grc-engineer: the core automation hub for gap assessment, IaC scanning, evidence collection, remediation generation, policy generation, PR review, continuous monitoring, and multi-framework optimization.
• Persona plugins: workflows for auditors, internal GRC teams, third-party risk, reporting, learning, and iterative GRC automation.
• Framework plugins: reference guidance for SOC 2, NIST 800-53, ISO 27001, FedRAMP, PCI DSS, CMMC, HITRUST, CIS Controls, GDPR, DORA, HIPAA Security, regional privacy/security regimes, and more.
• Connector plugins: thin wrappers around tools such as AWS CLI, GitHub CLI, gcloud, Azure CLI, Okta, Slack, Datadog, CrowdStrike, Drata, Splunk, Tenable, Snowflake, and POA&M automation.
• Diagram plugins: editable draw.io system boundaries, evidence flows, control maps, risk treatment, audit workflows, framework crosswalks, TPRM, POA&M, data flows, RACI, and operating model visuals.
• OSCAL and bridge plugins: tooling for FedRAMP/OSCAL workflows and integrations with external GRC systems.

The common path is:

connectors collect evidence
        ↓
findings match schemas/finding.schema.json
        ↓
grc-engineer maps findings through SCF
        ↓
reports, remediation, evidence packages, OSCAL outputs

The Secure Controls Framework (SCF) crosswalk is used as the control backbone: 1,468 controls mapped to 249 frameworks. The toolkit references control IDs and implementation guidance; it does not reproduce copyrighted standards text.

Install in 60 seconds

Inside Claude Code:

/plugin marketplace add GRCEngClub/claude-grc-engineering
/plugin install grc-engineer@grc-engineering-suite

For a first run without cloud credentials, use GitHub as the evidence source:

/plugin install github-inspector@grc-engineering-suite
/plugin install soc2@grc-engineering-suite
/github-inspector:setup
/github-inspector:collect --scope=@me
/grc-engineer:gap-assessment SOC2 --sources=github-inspector

Full walkthrough: docs/QUICKSTART.md.

Using Claude Desktop or Claude Cowork instead of Claude Code? Start with docs/CLAUDE-COWORK.md. Anthropic's security and compliance posture is documented at trust.anthropic.com, and the Claude Cowork third-party platform guide is here: Claude Desktop on third-party platforms.

Common workflows

Every command has a reference page in its plugin's commands/ directory.

Plugin map

Use /grc-engineer:frameworks to discover framework coverage and plugin depth.

High-level categories:

The marketplace manifest lives at .claude-plugin/marketplace.json.

Data contract

Every connector emits Findings that match schemas/finding.schema.json. A Finding is one resource with one or more control evaluations.

That contract keeps connectors small: each connector only needs to collect and normalize evidence. grc-engineer handles framework expansion, reporting, remediation, and downstream workflows.

For the full architecture and schema example, see docs/ARCHITECTURE.md.

Documentation

• docs/QUICKSTART.md: first gap assessment in about 10 minutes
• docs/ARCHITECTURE.md: pipeline model, data contract, plugin categories, extensibility
• docs/FRAMEWORK-COVERAGE.md: generated coverage for SCF-mapped frameworks
• docs/FRAMEWORK-PLUGIN-GUIDE.md: scaffold and level up framework plugins
• docs/GRC-DATA.md: user-owned grc-data/ contracts for metrics, risks, vendors, exceptions, and policies
• docs/ENTERPRISE-DEPLOYMENT.md: AWS Bedrock, Claude Platform on AWS, and Google Vertex AI guidance
• docs/CLAUDE-COWORK.md: Claude Desktop and Claude Cowork file-oriented usage, including third-party platform handoff notes
• Anthropic Trust Center: Anthropic security, compliance, and trust resources
• Claude Desktop on third-party platforms: official Cowork platform guidance
• docs/CONTRIBUTING.md: how to contribute connectors, framework guidance, and docs
• docs/SCF-ATTRIBUTION.md: SCF licensing and attribution

Contributing

Contributions are welcome from GRC practitioners, auditors, security engineers, platform teams, framework experts, and commercial GRC vendors.

The highest-value contributions are:

• New connector plugins
• Improvements to existing connectors
• Framework plugin guidance and evidence patterns
• Real-world remediation examples
• Documentation that helps practitioners learn GRC engineering

First-time contributors follow the issue → vouch → PR flow described in docs/CONTRIBUTING.md and GOVERNANCE.md.

Security-sensitive reports should use the private advisory process described in SECURITY.md.

Status

Pre-1.0. The Finding schema is versioned, and breaking changes are documented in CHANGELOG.md.

Star History

License

MIT for original code, copyright © GRC Engineering Club contributors. Exceptions are documented in LICENSE. The CIS Controls plugin is CC BY-SA 4.0 per upstream terms. SCF data is CC BY-ND 4.0 and redistributed verbatim.