Give your AI agent deep Bazel expertise — backed by real build data from Hermetiq.
This plugin adds a skill and MCP server that let Claude Code, Claude Desktop, or Codex analyze build performance, fix cache misses, optimize remote execution, and debug Buildbarn infrastructure using your project's actual metrics.
.bazelrc flags, hermeticity, and build graph patternsA Hermetiq account with at least one project receiving BEP events.
# Add the Hermetiq marketplace
/plugin marketplace add Hermetiq/hermetiq-ai-plugin
# Install the plugin
/plugin install hermetiq@hermetiq
The plugin marketplace is a Claude Code (CLI) feature, not Claude Desktop.
The plugin configures the MCP server automatically. On first use you'll authenticate via your Hermetiq account. Log in to dashboard.hermetiq.io beforehand for a seamless experience.
Authentication uses OAuth2 Dynamic Client Registration via Stytch. If you hit auth errors, re-authenticate with /mcp and follow the prompts.
To configure the MCP server without the plugin, add this to your Claude Code MCP settings:
{
"mcpServers": {
"hermetiq": {
"type": "http",
"url": "https://mcp.cloud-usc1.hermetiq.io"
}
}
}
Claude activates the skill automatically when you ask about build performance. You can also invoke it directly:
/hermetiq analyze my latest build
/hermetiq why is my cache hit rate dropping?
/hermetiq compare this week's builds to last week
Or ask naturally:
Why did my build take 20 minutes today?
Show me cache miss trends for the last 7 days
Which targets are most expensive to execute remotely?
Go to Settings > Connectors and add:
https://mcp.cloud-usc1.hermetiq.io
git clone https://github.com/Hermetiq/hermetiq-ai-plugin
cd hermetiq-ai-plugin/plugins/hermetiq/skills
zip -r hermetiq.zip hermetiq/
The zip must preserve the folder structure — it should contain
hermetiq/SKILL.md, not a bareSKILL.mdat the root.
Then in Claude Desktop:
hermetiq.zipGo to Settings > Capabilities and toggle on Code execution and file creation.
Skills require a Pro, Max, Team, or Enterprise plan.
codex mcp add hermetiq --url https://mcp.cloud-usc1.hermetiq.io
Verify with codex mcp list or codex mcp get hermetiq.
In Codex Desktop, click the Authenticate button. Or from the CLI:
codex mcp login hermetiq
Log in to dashboard.hermetiq.io first for a smoother flow. See Codex MCP docs for details.
git clone https://github.com/Hermetiq/hermetiq-ai-plugin
cd hermetiq-ai-plugin
mkdir -p ~/.codex/skills/hermetiq
cp -R plugins/hermetiq/skills/hermetiq/* ~/.codex/skills/hermetiq/
Restart Codex to pick up the new MCP server and skill.
Then ask build-performance questions naturally or invoke the skill by name:
Use the hermetiq skill to analyze my latest build
Why is remote execution queue time high this week?
Show cache miss reasons by mnemonic for my failing builds
| File | Description |
|---|---|
SKILL.md | Core skill with playbooks for build analysis, cache optimization, RBE tuning, and infrastructure debugging |
references/REFERENCE.md | Hermetiq data model, MCP tool inventory, and build configuration reference |
references/BUILDBARN_TUNING.md | Buildbarn infrastructure tuning guide for storage, workers, and scheduler |
External network access
Connects to servers outside your machine
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.
npx claudepluginhub hermetiq/hermetiq-ai-plugin --plugin hermetiqWrite SQL, explore datasets, and generate insights faster. Build visualizations and dashboards, and turn raw data into clear stories for stakeholders.
Give your AI a memory — mine projects and conversations into a searchable palace. 33 MCP tools, auto-save hooks, and guided setup.
MCP server that saves 98% of your context window with session continuity. Sandboxed code execution in 11 languages, FTS5 knowledge base with BM25 ranking, and automatic state restore across compactions.
Open-source, local-first Claude Code plugin for token reduction, context compression, and cost optimization using hybrid RAG retrieval (BM25 + vector search), reranking, AST-aware chunking, and compact context packets.
Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub
Excalidraw diagramming toolkit — auto-diagram any codebase, architecture diagrams, data flows, with PNG/SVG/URL export