Add CorpusIQ Hermes Community Hub — 302+ resources#505
Conversation
📝 WalkthroughWalkthrough
ChangesREADME Link Additions
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~2 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment Warning |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@README.md`:
- Line 93: The CorpusIQ Hermes Community Hub bullet entry has been incorrectly
added to four semantically incompatible sections of the README. Remove the
identical bullet entry from the Security Notice section (line 182), the Table of
Contents section (line 200), the Contributing section (line 1233), and the
License section (line 1247). Keep only the entry in the OpenClaw Ecosystem Tools
section (line 93) where it semantically belongs as a resource listing.
- Line 93: The CorpusIQ Hermes Community Hub entry should not be listed under
the OpenClaw Ecosystem Tools section since Hermes is a separate AI agent system,
not part of OpenClaw's native ecosystem. Either completely remove the line
containing the CorpusIQ Hermes Community Hub reference (the entry linking to
https://github.com/CorpusIQ/corpusiq-docs/tree/main/hermes), or relocate it to a
separate section like Third-Party Integrations with clarified context explaining
that it documents integration between Hermes and CorpusIQ's MCP server rather
than OpenClaw ecosystem components.
- Line 200: Remove the CorpusIQ Hermes Community Hub entry from the Table of
Contents in the README.md file. This entry breaks the established TOC structure
and format, which is designed as a grid of category links in the pattern
[Category Name](`#anchor-slug`) (count) that map exclusively to h3 headings within
details/summary blocks. The CorpusIQ entry is a standalone resource link with no
corresponding h3 section anchor in the document, making it orphaned and
inconsistent with the TOC's design pattern. Delete line 200 entirely to maintain
the structural integrity of the Table of Contents.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
|
|
||
|
|
||
| ## OpenClaw Ecosystem Tools | ||
| - [CorpusIQ Hermes Community Hub](https://github.com/CorpusIQ/corpusiq-docs/tree/main/hermes) — 302+ Hermes tools, skills, MCP servers, agents, blueprints, and case studies. |
There was a problem hiding this comment.
Questionable placement: The CorpusIQ Hermes Community Hub entry appears in five semantically incompatible sections.
The identical bullet entry is added to "OpenClaw Ecosystem Tools" (line 93—appropriate), but also to "Security Notice" (182), "Table of Contents" (200), "Contributing" (1233), and "License" (1247). Each of these latter sections serves a distinct purpose:
- Line 182 (Security Notice): This section disclaims that skills are curated but not audited—it's not a resource list.
- Line 200 (Table of Contents): This is an index of skill categories (Git & GitHub, Coding Agents, etc.), not individual tools. Adding a resource entry here breaks the TOC structure.
- Line 1233 (Contributing): This outlines submission and review guidelines, not a tool registry.
- Line 1247 (License): This is the legal/copyright notice, not a resource list.
The pattern suggests a copy-paste error across multiple hunks. Only line 93 (OpenClaw Ecosystem Tools) is the semantically correct location. If the hub belongs elsewhere, each placement needs justification specific to that section's purpose.
Also applies to: 182-182, 200-200, 1233-1233, 1247-1247
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@README.md` at line 93, The CorpusIQ Hermes Community Hub bullet entry has
been incorrectly added to four semantically incompatible sections of the README.
Remove the identical bullet entry from the Security Notice section (line 182),
the Table of Contents section (line 200), the Contributing section (line 1233),
and the License section (line 1247). Keep only the entry in the OpenClaw
Ecosystem Tools section (line 93) where it semantically belongs as a resource
listing.
🧩 Analysis chain
🏁 Script executed:
# First, let's examine the README.md file and get context around line 93
git ls-files | grep -i readmeRepository: VoltAgent/awesome-openclaw-skills
Length of output: 87
🏁 Script executed:
# Once we find the README, let's look at the context around line 93
git ls-files | head -20Repository: VoltAgent/awesome-openclaw-skills
Length of output: 659
🏁 Script executed:
# Search for README.md
fd -t f -i readme.mdRepository: VoltAgent/awesome-openclaw-skills
Length of output: 87
🌐 Web query:
CorpusIQ Hermes Community Hub GitHub repository
💡 Result:
There is no single repository officially named "CorpusIQ Hermes Community Hub." The terminology appears to be a conflation of two distinct parts of the AI agent ecosystem: 1. CorpusIQ: A company and platform that provides a business data layer for AI agents via an MCP (Model Context Protocol) server [1][2]. They maintain documentation at github.com/CorpusIQ/corpusiq-docs [3][4], but their core MCP server is a closed-source SaaS product [5]. 2. Hermes Agent Ecosystem: The Hermes Agent, developed by Nous Research, has a vibrant community that uses various hubs for sharing skills and tools [6][7][8]. Notable repositories and platforms in this ecosystem include: - HermesHub (github.com/amanning3390/hermeshub): A community-curated skills registry for Hermes Agent [9]. - Hermes Atlas (github.com/srinivasmd/hermes-ecosystem): A community-curated map and directory of tools, skills, and integrations for Hermes Agent [6][8]. Users in the community frequently combine these technologies—using Hermes agents to interact with CorpusIQ's MCP server to access business data [5][1]. While you may find discussions or "show & tell" issues on GitHub (such as issue #33441 in the NousResearch/hermes-agent repository) detailing how to connect these systems [5], there is no combined "CorpusIQ Hermes Community Hub" repository.
Citations:
- 1: https://www.corpusiq.io/mcp/connect-agent-to-mcp-server
- 2: https://www.corpusiq.io/
- 3: https://github.com/CorpusIQ/corpusiq-docs
- 4: https://trendshift.io/repositories/46991
- 5: Show & tell — building CorpusIQ MCP with Hermes, and using Hermes against CorpusIQ MCP NousResearch/hermes-agent#33441
- 6: https://github.com/srinivasmd/hermes-ecosystem
- 7: https://github.com/NousResearch/hermes-agent?aid=recVVCtCpQVPkq84K
- 8: https://get-hermes.ai/community/
- 9: https://github.com/amanning3390/hermeshub
🌐 Web query:
OpenClaw ecosystem tools MCP servers
💡 Result:
The OpenClaw ecosystem is a developer-focused AI automation platform designed for building, running, and orchestrating autonomous agents [1][2][3]. It acts as a local-first control plane for managing agent sessions, tools, channels, and persistent memory [4][5]. Ecosystem Tools The OpenClaw ecosystem includes several core components for managing the agent lifecycle [1]: - ClawHub: A registry for discovering, installing, and managing reusable skills (tools/plugins) for agents [1]. - ClawTasks: A built-in scheduler and automation layer for executing recurring jobs, multi-step workflows, and maintenance tasks [1][6]. - Moltbook: A conversation notebook used for capturing, tagging, annotating, and replaying agent-user interactions [1]. - Moltworker: A background worker responsible for executing slow or batch-oriented jobs [1]. - Unbrowse: A tool providing agents with controlled access to external web information [1]. - ClawI: A companion tool for observability, often used alongside Moltbook to inspect traces and agent behavior [1]. MCP (Model Context Protocol) Integration OpenClaw has robust support for the Model Context Protocol, which allows it to function as both an MCP server and an MCP client [7][8]: 1. OpenClaw as an MCP Server (openclaw mcp serve): When you run this command, OpenClaw acts as an MCP server using stdio transport [7][9]. This allows external MCP clients (such as Claude Code or other IDEs) to interface directly with your OpenClaw Gateway [7][8]. It bridges your routed channel conversations (e.g., Telegram, Discord, Slack) into MCP, exposing them as conversations and tools to the client [7][9][8]. 2. OpenClaw as an MCP Client Registry: OpenClaw manages outbound MCP server definitions in its configuration (stored in mcp.servers) [7][8]. Using commands like openclaw mcp add, set, and configure, you can define third-party MCP servers that your OpenClaw agents can consume [7][9][8]. Community MCP Servers: Several community-developed MCP servers exist to bridge OpenClaw with other tools [10][11][12]: - openclaw-mcp: A community-developed server designed to provide a secure bridge between Claude.ai and self-hosted OpenClaw assistants, supporting both synchronous chat and asynchronous long-running tasks [10]. - Helms-AI/openclaw-mcp-server: Exposes OpenClaw Gateway tools (messaging, scheduling, nodes, memory, etc.) to MCP-compatible clients like Claude Code [11]. - rodgco/openclaw-mcp-server: Exposes OpenClaw assistants to MCP clients (like Claude Desktop) using a Streamable HTTP transport with Server-Sent Events (SSE) for real-time streaming [12].
Citations:
- 1: https://clawsindex.com/ecosystem/
- 2: https://openclaw.im/
- 3: https://openclawai.io/solutions/developers
- 4: https://github.com/openclaw/openclaw
- 5: https://openclaw.ai/
- 6: https://clawsindex.com/ecosystem/clawtasks/
- 7: https://docs.openclaw.ai/cli/mcp.md
- 8: https://openclaw-ai.online/mcp/
- 9: https://docs.openclaw.ai/cli/mcp
- 10: https://explore.market.dev/ecosystems/openclaw/projects/openclaw-mcp
- 11: https://github.com/Helms-AI/openclaw-mcp-server
- 12: https://github.com/rodgco/openclaw-mcp-server
🏁 Script executed:
# Read the README.md file around line 93 to see the context
sed -n '80,110p' README.mdRepository: VoltAgent/awesome-openclaw-skills
Length of output: 1961
🏁 Script executed:
# Check if the URL path exists by examining what branches exist in CorpusIQ docs
# Let's also search for any reference to Hermes in the README
rg -i "hermes" README.md -A 2 -B 2Repository: VoltAgent/awesome-openclaw-skills
Length of output: 1400
🌐 Web query:
CorpusIQ/corpusiq-docs/tree/main/hermes GitHub path existence
💡 Result:
The GitHub path "CorpusIQ/corpusiq-docs/tree/main/hermes" exists and is publicly referenced as a documentation and configuration resource for integrating Hermes agents with the CorpusIQ Model Context Protocol (MCP) server [1][2]. This path provides guidance for developers on authentication and setup in non-interactive, agentic environments [3].
Citations:
- 1: https://www.linkedin.com/posts/benoit-pecqueur_hermes-aiagents-agenticai-activity-7470166507054858240-PIhv
- 2: https://www.linkedin.com/pulse/hermes-agent-101-natasha-lee-kaa5c
- 3: https://www.corpusiq.io/mcp/connect-agent-to-mcp-server
🌐 Web query:
Hermes agent Nous Research vs OpenClaw difference
💡 Result:
Hermes Agent and OpenClaw are both prominent, open-source AI agent systems, but they are built with fundamentally different architectural philosophies and primary objectives [1][2]. The core difference lies in what each system treats as its load-bearing element: Hermes Agent (by Nous Research) is agent-first, centered on a "closed learning loop" [3][4][2]. Its defining feature is its ability to learn and improve over time [1]. It automatically creates skills from completed tasks, refines those skills during subsequent usage, and maintains persistent memory to build a deepening model of the user across sessions [3][4][5]. It is designed as a background worker that grows more capable the longer it runs [1][2]. OpenClaw is gateway-first, centered on a "control plane" architecture [6][2]. It acts as a robust messaging hub, prioritizing reach and connectivity across 25+ platforms (e.g., WhatsApp, Discord, Slack, iMessage) [7][6][5]. Its primary value is infrastructure—routing messages, managing multi-agent environments, and providing granular access controls [2][5]. Rather than learning through persistent self-improvement, OpenClaw relies on a massive ecosystem of community-built plugins (via ClawHub) to provide breadth of functionality [8][5]. Key Comparison Table: Category | Hermes Agent | OpenClaw --- | --- | --- Core Philosophy | Learning Loop (Self-improvement) [2][5] | Gateway/Control Plane (Routing) [2][5] Primary Strength | Compounding intelligence over time [8][5] | Ecosystem reach and multi-platform support [1][8] Architecture | Agent-centric [2] | Infrastructure/Gateway-centric [2] Learning Style | Persistent/Compounding [2][5] | Session-native (no persistent learning) [8] Best Use Case | Specialist tasks that require deepening context [2][5] | Broad workflow orchestration & messaging [2][5] Because both systems are protocol-compatible and support standard interfaces like the Model Context Protocol (MCP), they are not mutually exclusive [5]. Many power users choose to run OpenClaw as the infrastructure gateway to handle routing, channels, and permissions, while nesting Hermes agents within that framework to execute tasks that require deep reasoning and persistent learning [2][5].
Citations:
- 1: https://www.hostinger.com/ca/tutorials/hermes-agent-vs-openclaw
- 2: https://dev.to/sebastian_chedal/hermes-agent-vs-openclaw-when-to-use-which-and-when-to-use-both-6e5
- 3: https://hermes-agent.nousresearch.com/docs/
- 4: https://github.com/NousResearch/hermes-agent?tab=readme-ov-file
- 5: https://cognio.so/resources/guides/openclaw-vs-hermes
- 6: https://docs.openclaw.ai/
- 7: https://github.com/OpenClaw/openclaw
- 8: https://lushbinary.com/blog/hermes-agent-vs-openclaw-updated-comparison-may-2026/
Remove or relocate the CorpusIQ Hermes Community Hub entry—it is not an OpenClaw ecosystem tool.
The URL is valid, but this resource should not be listed under "OpenClaw Ecosystem Tools." Hermes Agent (by Nous Research) and OpenClaw are fundamentally different AI agent systems with separate architectural designs: Hermes is agent-first and learning-focused, while OpenClaw is gateway-first and infrastructure-focused. The CorpusIQ Hermes Hub documents integration between Hermes and CorpusIQ's MCP server—two third-party systems—not OpenClaw ecosystem components. Including it in this section misrepresents the scope of OpenClaw's native ecosystem. Either remove it or move it to a separate section like "Third-Party Integrations" with clarified context.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@README.md` at line 93, The CorpusIQ Hermes Community Hub entry should not be
listed under the OpenClaw Ecosystem Tools section since Hermes is a separate AI
agent system, not part of OpenClaw's native ecosystem. Either completely remove
the line containing the CorpusIQ Hermes Community Hub reference (the entry
linking to https://github.com/CorpusIQ/corpusiq-docs/tree/main/hermes), or
relocate it to a separate section like Third-Party Integrations with clarified
context explaining that it documents integration between Hermes and CorpusIQ's
MCP server rather than OpenClaw ecosystem components.
|
|
||
|
|
||
| ## Table of Contents | ||
| - [CorpusIQ Hermes Community Hub](https://github.com/CorpusIQ/corpusiq-docs/tree/main/hermes) — 302+ Hermes tools, skills, MCP servers, agents, blueprints, and case studies. |
There was a problem hiding this comment.
Remove from Table of Contents; it breaks the TOC structure and format.
Lines 200–214 define the TOC as a grid of category links with skill counts: [Category Name](#anchor-slug) (count). The CorpusIQ entry at line 200 is neither a category link nor anchored to any corresponding h3 section in the document. It is orphaned and inconsistent with the TOC's design pattern, which maps exclusively to h3 headings within details/summary blocks (e.g., "Git & GitHub" → <summary><h3>Git & GitHub</h3></summary>).
Including a standalone resource link in the TOC creates visual/navigational confusion and violates the stated structure. Remove line 200 entirely from the TOC.
Also applies to: 202-214
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@README.md` at line 200, Remove the CorpusIQ Hermes Community Hub entry from
the Table of Contents in the README.md file. This entry breaks the established
TOC structure and format, which is designed as a grid of category links in the
pattern [Category Name](`#anchor-slug`) (count) that map exclusively to h3
headings within details/summary blocks. The CorpusIQ entry is a standalone
resource link with no corresponding h3 section anchor in the document, making it
orphaned and inconsistent with the TOC's design pattern. Delete line 200
entirely to maintain the structural integrity of the Table of Contents.
2 participants
Added the CorpusIQ Hermes Community Hub — the largest organized Hermes Agent knowledge base: 302+ tools, skills, MCP servers, agents, blueprints, and case studies across 18 categories. Built by heavy Hermes users. Updated nightly.
Summary by CodeRabbit