Runs the public HeyClaude MCP endpoint through mcp-remote when installed through stdio clients., Exposes read-only registry and draft-validation tools; it does not publish content, open PRs, or mutate GitHub.
Privacy notes
MCP queries and draft metadata are sent to the HeyClaude MCP endpoint when used through a remote client., Do not include credentials, private repository details, customer data, or unpublished material in draft fields.
Author
HeyClaude Team
Claim status
unclaimed
Last verified
2025-11-25
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
78
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Complete
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
Registry metadata indicates a reviewed listing.
Done
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
The official HeyClaude MCP server exposes the public HeyClaude registry to MCP
clients. It is designed for read-only discovery: search registry entries, fetch
details, inspect platform compatibility, copy install guidance, discover feeds,
and prepare reviewable content submissions.
The server does not publish content, create GitHub issues, open pull requests,
manage accounts, write local project files, or mutate editor configuration.
Submission helpers validate drafts and generate prefilled HeyClaude submit URLs
for a human-reviewed PR-first workflow.
Tools
search_registry: search public registry entries by query, category, and platform.
get_entry_detail: fetch an entry detail payload by category and slug.
Use the hosted endpoint directly if your MCP client supports remote Streamable
HTTP servers.
https://heyclau.de/api/mcp
Local stdio from the repository
Use this when developing HeyClaude locally or testing generated registry
artifacts before deploy.
pnpm --filter @heyclaude/mcp start
Set HEYCLAUDE_DATA_DIR=/absolute/path/to/apps/web/public/data if you need to
point the server at a specific generated artifact directory.
Example prompts
Search HeyClaude for MCP servers related to GitHub.
Show install guidance for mcp/github-mcp-server.
Validate this HeyClaude content submission draft and build a PR-first submit URL.
Find possible duplicate entries before I submit this new Claude skill.
Security model
Read-only registry artifacts are the source of truth.
Hosted MCP requests are schema validated and rate limited.
Submission tools return validation reports and URLs only.
No GitHub tokens, OAuth grants, API keys, PR creation, or issue creation are exposed through MCP.
No local project files, .cursor, .claude, or MCP client config files are written by the server.
Accepted submissions remain maintainer reviewed before content is imported.
Troubleshooting
The client cannot connect to the hosted endpoint
Confirm the client supports Streamable HTTP MCP servers. If it only supports
stdio, use mcp-remote with https://heyclau.de/api/mcp.
Search returns no results
Try a broader query or pass only one filter at a time. Categories use HeyClaude
slugs such as agents, mcp, tools, skills, rules, commands, hooks,
guides, collections, and statuslines.
A submission draft is invalid
Call get_submission_schema for the target category, then pass the required
fields into validate_submission_draft. The MCP server will report missing
fields, unsafe URLs, likely duplicates, and the next review step.
I need to publish content automatically
That is intentionally outside the MCP v1 boundary. Use the generated submit URL
or PR draft, then let the maintainer-reviewed import workflow handle accepted
submissions.
Show that HeyClaude MCP Server is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/heyclaude-mcp)
How it compares
HeyClaude MCP Server side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
2 trust signals differ across this comparison (Source provenance, Submitter).
The official OpenSearch Project MCP server (opensearch-mcp-server-py) that lets AI assistants query and operate OpenSearch clusters — listing indices, reading mappings, running Query DSL and PPL searches, inspecting shards and cluster health — over stdio or streaming transports, with basic-auth, AWS IAM, and header-based authentication.
Remote MCP server for X and Twitter automation: tweet search, profile timelines, follower export, media workflows, webhooks, and confirmation-gated write actions.
Official CircleCI MCP server that lets LLMs query build and test failures, detect flaky tests, check pipeline status, fetch build logs, and validate config in CircleCI through natural language.
✓Runs the public HeyClaude MCP endpoint through mcp-remote when installed through stdio clients.
Exposes read-only registry and draft-validation tools; it does not publish content, open PRs, or mutate GitHub.
✓Core tools are read/query-oriented, but the default-enabled GenericOpenSearchApiTool can call ANY OpenSearch API endpoint with a custom path, method, and body — including write and delete operations within the credential's permissions.
Opt-in categories add mutating tools; for example the memory and agentic-memory tools can create and delete stored memories (including delete-by-query), and search-relevance tools can create and delete configurations, query sets, judgments, and experiments.
Access is bounded by the OpenSearch credential and cluster RBAC, not by tool naming — scope the user/role to least privilege and prefer a non-production cluster when an agent acts autonomously.
Dynamic connection parameters let an agent pass `opensearch_url` and auth values per tool call, so a single running server can be pointed at multiple clusters; restrict which endpoints and credentials are available to the agent.
In streaming (SSE/Streamable HTTP) mode the server listens on a network port; do not expose it on a public interface without authentication in front of it.
✓Use API keys with write confirmations enabled because posting, follower exports, media actions, and webhooks can affect public accounts.
✓The server runs locally via npx and is spawned by your MCP client; the published bin entrypoint is dist/index.js.
Several tools take write/execution actions on your CI — run_pipeline, rerun_workflow, and run_rollback_pipeline can trigger pipelines, re-run workflows, and roll back deployed component versions. Review tool calls before approving them.
In remote (HTTP/SSE) mode the server listens on a network port (default 8000); set REQUIRE_REQUEST_TOKEN=true to reject unauthenticated requests and avoid exposing it on untrusted networks.
MAX_MCP_OUTPUT_LENGTH (default 50000) bounds response size; large log/artifact pulls are truncated rather than streamed in full.
Privacy notes
✓MCP queries and draft metadata are sent to the HeyClaude MCP endpoint when used through a remote client.
Do not include credentials, private repository details, customer data, or unpublished material in draft fields.
✓The server connects to your OpenSearch cluster with the credentials you supply; index documents, mappings, and search results it returns are passed to the LLM/MCP client and can include sensitive or PII fields stored in your indices.
Cluster-introspection tools (shards, nodes, cluster state/health, allocation, hot threads, tasks) can expose infrastructure metadata such as node hosts, system metrics, and index settings.
Credentials are provided via environment variables or per-call parameters — basic-auth passwords, AWS IAM role ARNs, and AWS profiles/keys — so keep client config out of version control and restrict access to it.
Memory and agentic-memory tools persist agent-authored statements into OpenSearch; treat that stored content and any semantic enrichment as retained data.
✓Timelines, profile data, follower exports, media, webhook details, and API credentials may be sent through model context.
✓CIRCLECI_TOKEN is a personal API token that grants the LLM access to your CircleCI projects, build logs, test results, artifacts, and usage data — treat it as a secret and scope it appropriately.
Build failure logs, test output, and artifacts retrieved by the tools are passed to your LLM/model provider; avoid exposing pipelines containing sensitive secrets or data.
download_usage_api_data and find_underused_resource_classes surface organization-level CircleCI usage and resource metrics.
Telemetry is collected by default; set DISABLE_TELEMETRY=true to opt out.
Prerequisites
MCP client with Streamable HTTP support, or stdio support via mcp-remote
Network access to https://heyclau.de/api/mcp
Optional local checkout if running the stdio package directly from the repo
Python 3.10 or newer and the `uv`/`uvx` package manager (or `pip install opensearch-mcp-server-py`)
A reachable OpenSearch cluster — self-managed, Amazon OpenSearch Service, or OpenSearch Serverless
Credentials for the chosen auth method — basic auth (username/password), AWS IAM role/profile, or header-based auth
An MCP-compatible client (Claude Desktop, Claude Code, Cursor, Kiro, VS Code, etc.)
Xquik account and API key from https://dashboard.xquik.com/account
Connected X account before using write actions such as posts, DMs, profile updates, or follows
MCP client with stdio support through mcp-remote or direct Streamable HTTP support
Network access to https://xquik.com/mcp
Node.js >= 18.0.0
A CircleCI Personal API token from https://app.circleci.com/settings/user/tokens
An MCP-compatible client (Cursor, VS Code, Claude Desktop, Windsurf, Copilot, or similar)
CircleCI projects you follow / have access to with the supplied token