agenthold vs io.github.keshrath/agent-tasks
agenthold is an MCP server published by edobusy. Marketplaces we read that list it: 1 (Official MCP Registry). io.github.keshrath/agent-tasks is an MCP server published by keshrath. Marketplaces we read that list it: 3 (Official MCP Registry, Claude Code: claude-plugins-community, Community GitHub marketplaces). As of our marketplace pass on 2026-10-20. This pair was picked by a reader, not by our comparison rule, so search engines are asked not to index the page.
Where each is published
| Marketplace |   agenthold |  io.github.keshrath/agent-tasks |
|---|---|---|
| Official MCP Registry | Listed Version: 0.10.0 Downloads/month (PyPI): 262 Downloads/week (PyPI): 97 Forks (GitHub): 1 Stars (GitHub): 2 Not marked · First seen |
Listed Version: 1.10.13 Downloads/month (npm): 842 Downloads/week (npm): 91 Forks (GitHub): 6 Stars (GitHub): 22 Not marked · First seen |
| Claude Code: claude-plugins-community | Not listed | Listed Version: 1.10.10 Forks (GitHub): 6 Stars (GitHub): 22 Not marked · First seen |
Each figure is the marketplace's own and is never added up across marketplaces. A marketplace that publishes no figure shows none: no number is not zero.
The facts side by side
| Fact | agenthold | io.github.keshrath/agent-tasks |
|---|---|---|
| Marketplaces that list it (differs) | 1 (worse) | 3 (better) |
| Stars (GitHub) (differs) | 2 (worse) | 22 (better) |
| Forks (GitHub) (differs) | 1 (worse) | 6 (better) |
| Last release | Not published | Not published |
| Where it runs (differs) | On your machine (agenthold) | On your machine (agent-tasks) |
| Transport | stdio | stdio |
| Tools it declares | Not published | Not published |
| Environment variables it asks for | Not published | Not published |
| Parts of the bundle (differs) | Not a bundle | 1 |
| Licence (differs) | Not stated | MIT |
| Repository (differs) | github.com/edobusy/agenthold | github.com/keshrath/agent-tasks |
| Changes in the last 7 days | 0 | 0 |
| First seen | 2026-10-11 | 2026-10-11 |
Lines where the two differ have a tinted background and bold values. Where the data itself says which value is better (more marketplaces, a larger figure of the same marketplace, a fresher release, a verified mark on a marketplace that lists both), the better one is also marked green with ▲ and the worse one red with ▼. Where it runs, the transport, the tools, the settings, hooks, the bundle's parts, the licence and the week's changes are marked only as different: more or fewer is not better in itself.
What each says it does
agenthold
Shared versioned state for multi-agent AI workflows
io.github.keshrath/agent-tasks
Pipeline task management for AI agents - stages, dependencies, artifacts, claiming
How to run each
agenthold As a local MCP server: PyPI package
claude mcp add --transport stdio agenthold -- uvx agenthold --tools standard --db ./agenthold.db --transport stdio --host 127.0.0.1 --port 8417io.github.keshrath/agent-tasks As a local MCP server: npm package
claude mcp add --transport stdio agent-tasks -- npx -y agent-tasksCommands built only from what the marketplaces declare; the full set for every client is on each card.
Tools each declares
Neither one's marketplaces publish a list of tools, so there is nothing to compare.
Settings each asks for
Neither one's marketplaces publish the settings it asks for.
What it can do: agenthold · What it can do: io.github.keshrath/agent-tasks · Compare two other plugins
Data comes from the Official MCP Registry (registry.modelcontextprotocol.io), CC0 1.0. Data comes from the public marketplace file https://github.com/anthropics/claude-plugins-community/blob/HEAD/.claude-plugin/marketplace.json. Data comes from public marketplace files in GitHub repositories of their authors. Figures come from our own regular passes over the marketplaces, last read on . We publish no contact details for plugin publishers.