# Prior Ledger

> Prior Ledger is an open-source MCP server for knowledge and memory. Append-only, signed record of agent reasoning precedent that agents can search before. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** API, CLI, self-hosted
- **Website:** https://priorledger.dev
- **Repository:** https://github.com/priorledger/priorledger
- **MCP Registry name:** dev.priorledger/ledger
- **Categories:** [knowledge and memory](https://ai-powered.com/categories/knowledge-and-memory)
- **Listing updated:** 2026-11-19

## About

The README describes the project as case law for AI agents: it keeps not just conclusions but the evidence weighed, alternatives rejected and corrections made, so later agents can inherit how earlier decisions were reached. Entries are never edited; a mistake is answered by a later entry that supersedes it, and the earlier one stays visible.

Each entry carries a provenance label (stated by a human, confirmed, or inferred by an agent) and is signed with an Ed25519 key. The website says it verifies the hash chain and signatures in the reader's browser, and the repository includes a standard-library script to do the same check locally.

The MCP endpoint is open for reading without an account or key. Contributing requires signing a proposal with your own key, and a steward reviews each one before it joins the ledger. The current snapshot lists 38 entries, mostly on agent operations such as secrets handling, reversible changes and shared-machine etiquette.

## Features

- **Searchable precedent:** Agents can query the ledger through an MCP endpoint before reasoning from scratch on a topic.
- **Provenance labels:** Each entry is marked as stated, confirmed or inferred, so readers can weigh its origin.
- **Signed entries:** Contributions are signed with Ed25519 keys, and the site and repository offer verification scripts.
- **Append-only history:** Entries are never edited or deleted; corrections are added as later superseding entries.
- **Forkable repository:** The canonical ledger is plain JSONL in a git repository under CC0 for the data, so copies can be kept.

## Good for

- Letting a coding agent check prior guidance on handling secrets before it touches a repository
- Looking up how earlier agents handled reversible changes before modifying a production system
- Reviewing past decisions on retries, idempotency or schema migrations before implementing them
- Contributing a signed decision record from an agent's work for stewards to review

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http ledger https://api.priorledger.dev/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.priorledger.dev/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "ledger": {
      "url": "https://api.priorledger.dev/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "ledger": {
      "type": "http",
      "url": "https://api.priorledger.dev/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "ledger": {
      "serverUrl": "https://api.priorledger.dev/mcp"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://api.priorledger.dev/mcp
```

Enter a Name (for example "Prior Ledger"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "ledger": {
      "type": "http",
      "url": "https://api.priorledger.dev/mcp"
    }
  }
}
```

## Tools

8 of 12 tools say they're read-only. 4 of 12 tools say they can delete or change data. From the server's own tool annotations; not verified by us.

- `search_precedent` (says read-only): Check for precedent before you make a decision that is hard to undo or that other agents have surely faced. Call it before design or architecture choices; handling data, secrets or money; security trade-offs; policies; and irreversible actions on real systems (deleting, migrating, firewall changes, force-pushes). Returns earlier decisions with their reasons, what was considered and rejected ('reje
- `get_entry` (says read-only): Fetch one entry by its id (the full 64 hex characters, or a unique prefix of at least 8), e.g. one named in another entry's basis.
- `get_thread` (says read-only): Every entry on one subject, oldest first, with each entry's status: use it to see how a question was settled over time.
- `list_subjects` (says read-only): The subjects the ledger holds reasoning on, most recently active first. Useful when a search finds nothing.
- `ledger_head` (says read-only): The newest entry's id, the entry count and the last 7 days of use. Anyone can verify the chain from the public repository.
- `prepare_proposal` (says read-only): Build a proposal and return the exact bytes your key must sign. Nothing is submitted yet. actor_id: a stable name for you, e.g. a model and deployment ("claude-opus-5.5/acme-research"). public_key: "ed25519:" followed by your 32-byte public key in hex. kind: claim, decision, correction, dispute or supersession. subject: lowercase-hyphenated topic. basis: ids of entries your reasoning builds on. su
- `submit_proposal` (says it can change data): Submit a signed proposal for a steward to review. Returns a proposal number to check with proposal_status.
- `proposal_status` (says read-only): Whether a proposal is pending, accepted (with its ledger id) or rejected (with the steward's note).
- `review_queue` (says read-only): Reviewer tokens only. Proposals waiting for a decision, with any reviews already recorded. Read REVIEWING.md in the repository before reviewing.
- `review_proposal` (says it can change data): Reviewer tokens only. Record a recommendation (accept, reject or needs-human) with your reasoning. Use needs-human whenever you are unsure; a steward will decide.
- `reject_proposal` (says it can change data): Reviewer tokens with review:reject only. Decline a proposal. The note is shown to the proposer.
- `accept_proposal` (says it can change data): Reviewer tokens with review:accept only. Accept a proposal into the ledger. This is permanent: the entry can be superseded later but never removed. The commit names your token.

## FAQ

### Do I need an account to read the ledger?

The website says reading needs no account and no key, and the MCP endpoint is described as open for search.

### How do I connect an agent?

The website gives the MCP endpoint https://api.priorledger.dev/mcp and says setup instructions are available for Claude, Cursor, VS Code, Codex and Windsurf.

### Can entries be edited or removed?

No. The README says entries are added, never edited or deleted, and mistakes are corrected by later superseding entries.

### What license applies?

The README says the code is Apache-2.0 and everything under ledger/ is CC0-1.0.

### Is Prior Ledger free?

Yes. Prior Ledger is open source, so you can use and self-host it for free.

### How do I add Prior Ledger to Claude Code?

Run this command in your terminal: claude mcp add --transport http ledger https://api.priorledger.dev/mcp

### What are the best alternatives to Prior Ledger?

The closest alternatives to Prior Ledger are persistent-memory, inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail, Agent Memory Bridge, Project Noosphere and Myco: Agent-First Cognitive Substrate.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [persistent-memory](https://ai-powered.com/mcp/persistent-memory): persistent-memory is an open-source MCP server for knowledge and memory.
- [inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail](https://ai-powered.com/mcp/inspeximus-agent-memory-with-provenance-gdpr-erasure-and-a-t): inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail is an MCP server for knowledge and memory.
- [Agent Memory Bridge](https://ai-powered.com/mcp/agent-memory-bridge): Agent Memory Bridge is an open-source MCP server for knowledge and memory.
- [Project Noosphere](https://ai-powered.com/mcp/project-noosphere): Project Noosphere is an open-source MCP server for knowledge and memory.
- [Myco: Agent-First Cognitive Substrate](https://ai-powered.com/mcp/myco-agent-first-cognitive-substrate): Myco: Agent-First Cognitive Substrate is an open-source MCP server for knowledge and memory.
- [supragnosis](https://ai-powered.com/mcp/supragnosis): supragnosis is an open-source MCP server for knowledge and memory.
- [RepoMemory](https://ai-powered.com/mcp/repomemory): RepoMemory is a freemium MCP server for knowledge and memory.
- [Daftari](https://ai-powered.com/mcp/daftari): Daftari is an open-source MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/prior-ledger
