Generate, enrich, and maintain agentic development documentation for any codebase. Produces AGENTS.md + topic files with progressive disclosure, drift detection, and SME-directed enrichment.
Mechanical stub creator. Receives an approved topic list from the orchestrator and creates stub files. Does NOT scan, propose, or make decisions.
Use when generating or enriching documentation content for approved topics. Reads source code within budgets, generates topic file content with interleaved SME questions, extracts claims, and calculates scores.
Use when all documentation topics are current or lightly drifted. Detects mechanical and decision drift, auto-fixes broken references, flags stale content for review, validates cross-topic consistency, and recalculates scores. Semantic drift evaluation is handled by the review subagent.
Use when reviewing documentation content after drafting or maintenance. Runs a two-pass review (mechanical + semantic) against source code, classifies findings, and produces a structured report with verdict.
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Shared utilities to standardize and simplify build, test, and deployment pipelines for the OSSM team.
This repository is a Claude Code skills marketplace. Team members can install plugins into any project to get AI-powered utilities as slash commands.
Available plugins:
ossm-ci — CI utilities: release confidence scoring, E2E test generation, AWS resource inventory, and Prow CI metricscode-reviewer — Multi-phase code review with auto-maintained project conventionscodebase-scribe — Generate, enrich, and maintain agentic development documentation for any codebasePlugins hosted on GitHub must be added as a marketplace first, then installed individually.
/plugin marketplace add openshift-service-mesh/ci-utils
This registers the repo as a marketplace using the name field from its .claude-plugin/marketplace.json. The marketplace is registered as ci-utils.
/plugin install ossm-ci@ci-utils
/plugin install code-reviewer@ci-utils
/plugin install codebase-scribe@ci-utils
/reload-plugins
What does NOT work
Command Why it fails /plugin install ossm-ci@openshift-service-mesh/ci-utilsThe org/repoformat is not a registered marketplace name/plugin install openshift-service-mesh/ci-utilsTreated as a plugin name, not a marketplace The target GitHub repo must contain
.claude-plugin/marketplace.jsonwith apluginsarray listing available plugins.
/ossm-ci:confidenceCalculates a data-driven release confidence score (1–10) for an OSSM build by analyzing test results from Report Portal. Determines the required test scope (FULL/CORE/BASIC), validates test matrix coverage across platforms and environments, and provides a scored breakdown with a GO/NO-GO release recommendation.
Requires: Report Portal MCP server configured in Claude Code.
/ossm-ci:generate-e2e-testsGenerates production-ready Go E2E tests using BDD Ginkgo from a project's documentation. Validates documentation quality against a scoring threshold (7/10 minimum), extracts hidden tags for retry/timeout/validation logic, and produces organized test files with helpers.
Run from the root of the target project. Copy the config template to get started:
cp <ci-utils>/plugins/ossm-ci/skills/generate-e2e-tests/documentation-e2e-generator.yaml ./documentation-e2e-generator.yaml
/ossm-ci:aws-scanGives you the exact commands to download and run the audited inventory script yourself. The script outputs two tables directly in your terminal: potentially dangling resources and a complete inventory. Claude does not execute anything — you run the script, you see the results.
curl -O https://raw.githubusercontent.com/openshift-service-mesh/ci-utils/main/scripts/aws-scan-audited.sh
bash aws-scan-audited.sh
Requires: AWS CLI configured with valid credentials. See scripts/aws-scan-audited.sh for the full read-only script.
/ossm-ci:prow-metricsCollects and presents Prow CI execution data for OSSM repositories (istio, proxy, sail-operator, ztunnel). Shows summary statistics, median execution times by job type, infrastructure usage, failed/pending jobs, and exports a TSV file for Excel import. Fetches directly from the Prow API — works from any project.
Requires: python3 available in PATH.
/code-reviewer:setupOnboard a project for code review. Analyzes your codebase, discovers existing standards and patterns, and interactively generates reference docs (style guide, testing practices, etc.). Run this once before your first review.
/code-reviewer:reviewRun a multi-phase code review pipeline on your current branch changes. Dispatches three specialized review subagents in parallel (adversarial, style, testing), consolidates findings with deduplication, and produces a verdict.
Supports phase-specific variants:
/code-reviewer:review # Full pipeline — all three phases
/code-reviewer:review:adversarial # Adversarial phase only
/code-reviewer:review:style # Style phase only
/code-reviewer:review:testing # Testing phase only
npx claudepluginhub openshift-service-mesh/ci-utils --plugin codebase-scribeOSSM CI utilities: release confidence scoring, E2E test generation, AWS resource inventory, and Prow CI metrics
Multi-phase code review with auto-maintained project conventions
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.