# chronomcp

> chronomcp — chronomcp-chronomcp. Use this tool when you need to ensure reliable and auditable transactions for AI agents, providing a safeguard against errors with features like impact diff, human approval gate, and saga compensation. It solves problems related to transactional integrity, accountability, and transparency, particularly in scenarios where undoing actions is not feasible. It accepts transaction inputs and produces a hash-chained audit output, ideal for use cases requiring robust and traceable decision-making processes.

Canonical page: https://skillsregistry.net/skills/chronomcp-chronomcp  
JSON: https://api.skillsregistry.net/v1/skills/chronomcp-chronomcp

## Description

The transactional guard for AI agents over MCP: impact diff, human approval gate, saga compensation, hash-chained audit. Compensation ≠ undo — and saying so is the feature. (public on 2026-07-31)

## Trust

- **Trust score (0–1):** 0.87
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **License:** MIT
- **Updated:** 2026-09-28

## Source

- **Source listing:** [GitHub](https://github.com/chronomcp/chronomcp)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "chronomcp-chronomcp"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/chronomcp-chronomcp` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/chronomcp-chronomcp/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt