Signed, versioned ledger of standing instructions with intake contradiction diffs.
Scored across 6 tools
Each tool has a distinct role: creation, rule compilation, order diffing, execution, signature verification, and version/chain inspection. The overlap between ledger_diff and ledger_execute is intentional and clearly described as independent re-diffing during execution.
All tools use the same ledger_ prefix followed by a concise verb or noun with consistent snake_case. The pattern is predictable throughout.
Six tools form a tight, well-scoped set covering the core ledger lifecycle without redundant operations. Each tool earns its place.
The surface covers create, compile, diff, execute, verify, and version, which is strong for a signed ledger domain. Minor gaps exist around explicit read/list operations for ledger entries, but the core lifecycle is covered.
Compile a standing rule into a signed ledger entry (or a signed residual entry when it cannot compile faithfully).
Inputs: id (required), text (required), ledger_id (required), rank, +1 more.
Create a standing-instruction ledger under your account.
Inputs: name (required).
Diff an order against a ledger at intake: CONSISTENT / CONTRADICTS / APPROVAL-REQUIRED / EXPIRED-RULE-INVOKED / OVERRIDE_RECEIPT / CONFIRMATION-REQUIRED. Metered like the HTTP diff.
Inputs: order (required), ledger_id (required).
Execute an order: the server independently re-diffs (409 semantics as an error result on contradiction) and returns an attestation naming the ledger version in force.
Inputs: order (required), ledger_id (required).
Verify a ledger record's Ed25519 signature against the published keys.
Inputs: record (required).
A ledger's current version (head hash) and chain verification.
Inputs: ledger_id (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Append-only decisions with provenance, supersession, retrieval, and audited MCP actions.
The record of what the agent relied on: a signed, chained receipt per cleared read. Verify by id.
A handling-condition ledger that travels with a sensitive payload across every agent hop.
Signed, append-only receipts of an AI agent's decisions and outcomes, with public proof links.
Add one secure layer between your agents and this server.
