janus
v0.24.0OpenShift/RHEL/CNV research & investigation pipeline for Claude Code — crash forensics, upgrade/compat analysis, CVE impact, component behavior. Eight composable agents — doc-search (okp-mcp + mslearn for ARO/Azure + AWS docs for ROSA/AWS), source-trace (versioned-source index, optional), github-trace (github, conditional), jira-trace (mcp-atlassian, conditional), crash-analyze (drgn/gdb), iac-author (terraform/ansible MCP — authors and statically validates the lab's IaC, provisions nothing), lab-verify (oc/terraform/linux-mcp), synthesize — connected by a universal finding format, plus periodic self-improver and upstream-adviser. Reusable investigation patterns are inlined into each agent. Working-method discipline built in: evidence-basis labels (VERIFIED/REASONED/ASSUMED), named report-acceptance gates, a causation gate for crash findings, a project-local lessons loop, a deterministic secret-safety PreToolUse hook, and a tamper-evident per-case evidence chain (append-only hash-linked chain.jsonl, auto-sealed by a PostToolUse hook, verified at synthesis and handoff), a fan-in write-lock that freezes the fact base against mid-collaboration rewrites (chain.py lock + a PreToolUse deny hook), and mechanical reference-URL liveness and verbatim-quote fidelity checks behind the report gates, a validated tool-grant boundary keeping infrastructure provisioning behind the human approval gate, a Japanese report mode whose prose is checked mechanically with textlint (ja-technical-writing) behind gate C2/prose, and click-through evidence links from every report claim to the finding behind it, verified to resolve behind gate C1/link.
By nogunixLicense: MIT0 GitHub starsUpdated 2 hours ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 6 skill or MCP entries
- Source updated
- Aug 29, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install plugin
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install janus-2@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/nogunix/janusClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/janus/.
Plugin files
├── .claude-plugin/plugin.json├── skills/deck/SKILL.md├── skills/gslides/SKILL.md├── skills/janus/SKILL.md├── skills/md2pdf/SKILL.md├── skills/ocp-triage-heuristics/SKILL.md└── skills/okp-doc-search/SKILL.md
Included Skills6
Build a polished PowerPoint (.pptx) and PDF natively on top of an existing .pptx template — keeping the template's master, theme, layouts, logo, footer and brand colors while filling in your own content. Use when asked to create / generate slides or a slide deck / presentation from a template, rebuild a deck in a brand template, restyle content into a corporate template, recolor dark diagrams to match a light template, or convert pptx to PDF (incl. Japanese text that otherwise renders in Chinese glyphs). Linux + LibreOffice + python-pptx.
Create Google Slides presentations via gws CLI — declarative YAML spec or direct API calls. Supports text, tables, images, shapes, brand guidelines, SVG/raster diagram insertion with optional light-recolor. No pptx, no LibreOffice — the result is a live Google Slides document in Drive, ready for collaborative editing. Use when asked to create / generate slides, a deck, or a presentation.
Autonomous research & investigation pipeline for OpenShift / RHEL / OpenShift-Virtualization (CNV/KubeVirt). Case types: kernel crash & forensics (vmcore / panic / Oops / OOM / hung-task / guest crash, via drgn), upgrade & cross-version compatibility analysis, CVE/errata impact assessment, operator/component behavior investigation, release diffing. Correlates symptoms with CVE/errata/KB via okp (plus versioned-source tracing when a source-index server is registered), and produces a ranked-hypothesis report handed off to a human review queue. Read-only, sandboxed; never spoofs guardrails. Triggers: "vmcoreを解析", "このpanicの原因", "OOM調査", "アップグレード互換性を調査", "CVEの影響評価", "オペレータの挙動を調査", "バージョン間の差分を調査".
Convert Markdown to PDF with proper Japanese typography via pandoc + weasyprint. Use when asked to create / generate a PDF from a markdown file, convert md to PDF, or produce a printable document. Handles CJK fonts, tables, code blocks, SVG images, and mermaid diagrams. Triggers: "PDFにして", "PDF作成", "md to pdf", "ドキュメントをPDFに".
Experiential OpenShift live-cluster triage heuristics — layered triage, failure-mode classification, operator status-triple, node lifecycle, and upgrade gates. Distilled for JANUS from the openshift-ops plugin. Read-only diagnostics are autonomous-safe; every mutating remediation is flagged and stays behind JANUS human-approval gates.
How to research Red Hat documentation, KB solutions, CVEs, and errata effectively through okp-mcp (search_portal / get_document). Use whenever answering a Red Hat / RHEL / OpenShift / CNV question from official documentation, resolving an access.redhat.com/solutions URL or a docs.redhat.com URL, or when get_document returns "Document not found". Covers query construction, Solr doc_id path rules, URL-to-keyword extraction, and whole-guide mapping. Triggers: "Red Hatのドキュメントを 調べて", "このソリューション記事の内容を確認", "KB検索", "エラータを確認".
Plugin manifests1
{
"name": "janus",
"description": "OpenShift/RHEL/CNV research & investigation pipeline for Claude Code — crash forensics, upgrade/compat analysis, CVE impact, component behavior. Eight composable agents — doc-search (okp-mcp + mslearn for ARO/Azure + AWS docs for ROSA/AWS), source-trace (versioned-source index, optional), github-trace (github, conditional), jira-trace (mcp-atlassian, conditional), crash-analyze (drgn/gdb), iac-author (terraform/ansible MCP — authors and statically validates the lab's IaC, provisions nothing), lab-verify (oc/terraform/linux-mcp), synthesize — connected by a universal finding format, plus periodic self-improver and upstream-adviser. Reusable investigation patterns are inlined into each agent. Working-method discipline built in: evidence-basis labels (VERIFIED/REASONED/ASSUMED), named report-acceptance gates, a causation gate for crash findings, a project-local lessons loop, a deterministic secret-safety PreToolUse hook, and a tamper-evident per-case evidence chain (append-only hash-linked chain.jsonl, auto-sealed by a PostToolUse hook, verified at synthesis and handoff), a fan-in write-lock that freezes the fact base against mid-collaboration rewrites (chain.py lock + a PreToolUse deny hook), and mechanical reference-URL liveness and verbatim-quote fidelity checks behind the report gates, a validated tool-grant boundary keeping infrastructure provisioning behind the human approval gate, a Japanese report mode whose prose is checked mechanically with textlint (ja-technical-writing) behind gate C2/prose, and click-through evidence links from every report claim to the finding behind it, verified to resolve behind gate C1/link.",
"version": "0.24.0",
"author": {
"name": "nogunix"
},
"homepage": "https://github.com/nogunix/janus",
"license": "MIT"
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[janus on Agent Plugins Marketplace](https://pluginsmp.com/plugins/janus-2)