By RunemarkLabs
[RuneSmith v0.8.3] Developer helpers for the marketplace itself. /help lists installed skills and their natural-language triggers, /plugin-builder scaffolds new plugins, /tech-debt scans the workspace for stale or orphaned artifacts, and /skill-updater propagates convention changes across all skills consistently.
Show all available workspace skills, installed plugins, and what they do.
Create Cowork plugins and install them into Claude Desktop.
Propagate architectural decisions and convention changes across all skills in installed plugins and bootstrap repos.
Scan the workspace for stale artifacts, orphaned files, and technical debt.
Show all available workspace skills, installed plugins, and what they do. Use when the user says "help", "what can you do", "list skills", "show commands", "what skills are there", or seems unsure what's available. Also use when the user asks "how do I...", "is there a way to...", or "can you..." about a workflow that might already be covered by a plugin or skill.
Create Cowork plugins from scratch with full skill scaffolding, production-ready frontmatter, and audit-passing structure. Use when the user says "create a plugin", "new plugin", "add a skill", "make a plugin", "build a plugin", "I need a new skill", "scaffold a plugin", "start a plugin from scratch". Also triggers on "package this plugin" or "install this plugin". Not for editing existing skills (edit directly) or installing from catalog (use /core:install).
Propagate convention changes across skills safely by orchestrating per-skill modifications through Anthropic's skill-creator. Use after a decision changes how the system works - variable renames, workflow changes, lib reference updates, removed features, new conventions. Triggers on: "update skills", "propagate this", "skills are stale", "sync the change", "make sure skills reflect this", or when a decision contradicts what existing skills describe.
Scan the workspace for stale, orphaned, or unreferenced artifacts using a cross-reference graph (plans → refs/tickets/decisions → research/source-docs/drafts). Identifies what is ACTUALLY tech debt - content that no longer traces back to a live plan, ticket, or decision. Workspace-side only; never enters source repos. Use after a round of changes, when things feel messy, proactively between work sessions, or to verify cleanliness before a release. Triggers on: "tech debt", "clean up", "stale files", "what's out of date", "is anything orphaned".
Uses power tools
Uses Bash, Write, or Edit tools
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.
Claude plugin marketplace for planning, Atlassian (Jira + Confluence) integration, Claude Code workspace bootstrap, AI Operations documentation, and development tooling. Generic and single-tenant configurable per project - every tenant detail comes from user input or .credentials at runtime.
Built and maintained by Runemark Labs. Apache-2.0 licensed.
Eight plugins, ~30 skills total, ~30 commands. Install only what you need.
| Plugin | Skills | Purpose |
|---|---|---|
runesmith-core | plan, install, sync, setup | Foundation. Credentials, plugin management, chat-first planning. Required base. |
runesmith-workspace | reallocate, inbox | Workspace folder structure: _INBOX/ drop zone, plans/active/, plans/archive/, snapshots. |
runesmith-cc | bootstrap-cc, guardrail | Build the {PROJECT}.cc/ Claude Code monorepo head with comms folder; create or clone repos via PAT. Install the user-level project-boundary guardrail that constrains CC sessions to their launch project. |
runesmith-jira | ticket, bug-report, project-status, new-project | Jira tickets and project basics with current Cloud REST endpoints. |
runesmith-confluence | feature-doc, architecture-doc, project-overview, decisions-log, known-issues, roadmap, session-log | Confluence pages with storage XHTML conversion. |
runesmith-sprint | enable, disable, start-sprint, sprint-status, check-comms, plan-to-tickets | Sprint workflow + Cowork↔Claude Code interconnect via comms protocol. |
runesmith-aiops | bootstrap-aiops | Populate an AI Operations Confluence space from six template pages. |
runesmith-devtools | help, plugin-builder, tech-debt, skill-updater | Developer helpers for the marketplace itself. |
comms/, Claude Code executes. No Atlassian dependency./runesmith-sprint:enable and /runesmith-sprint:disable./plugin marketplace add runemarklabs/runesmith
/plugin install runesmith-core@runesmith
/plugin install runesmith-workspace@runesmith
/plugin install runesmith-cc@runesmith
/plugin install runesmith-jira@runesmith
/plugin install runesmith-confluence@runesmith
/plugin install runesmith-sprint@runesmith
/plugin install runesmith-aiops@runesmith
/plugin install runesmith-devtools@runesmith
After install, run /runesmith-core:setup to configure credentials. See INSTALL.md for full options.
.plugin files (Cowork desktop).plugin zips are not committed to this repo - they're generated on demand. Build them first:
python scripts/build.py # cross-platform; recommended on Windows
# or:
bash scripts/build.sh # Unix/Mac
This produces runesmith-*.plugin files in ./dist/ (gitignored). Drag those into the Cowork plugin sidebar to install.
Custom output directory:
python scripts/build.py path/to/output/
Every write skill follows the same gate:
to: user items first..credentials and project values./drafts/, /plans/active/, or chat.No write happens without the trigger. No skill talks to deprecated endpoints. No body is markdown.
Cowork ↔ Claude Code communicate via files in {PROJECT}.cc/comms/. CC writes ambiguity, blockers, or user-action requests to comms/open/; Cowork triages on /runesmith-sprint:check-comms or any planning interaction (check-on-entry pattern). User is reached only through Cowork - CC never asks the user directly.
npx claudepluginhub runemarklabs/runesmith --plugin runesmith-devtools[RuneSmith v0.8.0] Jira ticket and project workflows: create tickets (Task/Story/Bug/Epic) with proper ADF formatting, report bugs with linked Confluence docs, scaffold new Jira projects, and read board status. All operations run against current Atlassian Cloud REST endpoints (no deprecated paths) with explicit consent gates before any push.
[RuneSmith v0.8.0] Atlassian sprint workflow + Cowork↔CC interconnect. /enable wires Atlassian into a project (CLAUDE.md updates, CC-side skill deployment), /start-sprint and /sprint-status manage the active sprint handshake, /check-comms triages the comms folder, /plan-to-tickets converts plans into Jira tickets organized by sprint. CC reads sprints via read-only REST while mutations route through Cowork.
[RuneSmith v0.8.4] Builds the {PROJECT}.cc/ Claude Code monorepo head at your workspace root and installs the project-boundary guardrail that constrains Claude Code sessions to their launch project. /bootstrap-cc writes the CC head (CLAUDE.md, .claude/ scaffolding, comms folder, marker files) and project-level settings; /guardrail installs the user-level permission rules and PreToolUse hook that block cross-project filesystem reads. Idempotent - safe to re-run.
[RuneSmith v0.8.0] Populates an AI Operations Confluence space from six template pages (Quick Start, Full Integration, Architecture, Best Practices, FAQ, Reference). All canonical tokens ({COMPANY}, {SITE}, {SPACE_KEY}, {PROJECT_KEY}, etc.) are substituted at publish time so the same templates work for any tenant. Space must already exist in Confluence - skill is a one-shot populator.
[RuneSmith v0.8.0] Confluence page authoring with proper storage XHTML conversion. Skills cover feature specs, architecture decision records (ADRs), project overview landing pages, append-only decisions logs, known issues trackers, roadmap pages, and session logs. Drafts in markdown locally, converts and publishes on user consent. Optionally pulls plans/active/ as prefill source.
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.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Complete developer toolkit for Claude Code
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
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.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.