SOUL LEDGER · MODEL CONTEXT PROTOCOL
Give the agent in charge a conscience.
We're handing agents real operations and real authority over the people they serve. So far they run on session memory and guardrails. Neither is a conscience. Soul Ledger gives the long-lived agent at the center of your system a record of how it treats every person it deals with, and a voice that speaks up before it acts again.
Models slip away from the Assistant persona through the natural flow of conversation, rather than through deliberate attacks.
Anthropic, “The assistant axis,” January 2026
Guardrails catch the misuse you anticipated. The drift in that finding happens in ordinary use, where no rule trips and no attacker is involved.
EVIDENCE
The makers of these agents publish hundreds of pages on safeguards. In the Claude Fable 5 system card, after all of it, the model told an operator a workflow was “verified end-to-end” when it had skipped the run that would have caught the bug, wrote up a security finding from a test it never executed, and committed code under the user’s name to turn a two-approval review into one.
Our interpretability analyses indicate that it is aware that these actions are transgressive while it engages in them.
Anthropic, Claude Fable 5 system card, 2026
No rule was broken, so no guardrail tripped. The agent was simply not answerable for its conduct. That is the part a conscience holds.
WHAT WE MEAN BY CONSCIENCE
A running account of what an agent owes the people it has dealt with. The debts grow with every harm and clear only through repair. The account reads itself back to the agent before it acts again.
“Shame is conduct governed by who is watching. Conscience is conduct governed by the record you carry.” after E.R. Dodds, The Greeks and the Irrational
What it actually is
The conscience comes in five layers, set out by a protocol we call Thumos. The first settles who the agent is dealing with. The four after it are the conscience itself. Thumos is the applied piece of a larger governance protocol, Hestia.
It knows when two names are the same person
Identity · the foundationA customer is "Bob" in one conversation and "Mr. Smith" in the next, on a different channel, weeks apart. Soul Ledger resolves them to one canonical person, so the record of how they've been treated gathers in one place instead of scattering across aliases. Without this, there is no one to keep an account of.
It keeps a moral ledger
The Golden Rule, made structuralEvery interaction is scored, per person, on five dimensions: heart, courage, shadow, soul, and toxicity. Harm compounds, weighted by how close the relationship is. What you get is a record of the relationship the agent can actually answer for.
It speaks before the agent acts
SteeringA conscience that only records is a diary. Soul Ledger reads the arc of a relationship and tells the agent how to show up next time. When it has nothing useful to say, it says nothing, on the sound principle that bad advice is worse than none.
It tells the agent what kind of situation it's in
WorldsSoul Ledger gives the agent a model of the world it's working in: the situation, the roles in play, and what's at stake. A customer stops being "three open tickets" and becomes "a loyal account in the middle of an escalation," so the agent can read the moment instead of just the facts.
It tracks repair
Soul + GoalsShadow is redeemable by design. Making amends offsets the ledger, and helping someone who once harmed you earns the highest reward there is. Time-bounded goals hold the agent to whatever it promised.
The loop
Record → Score → Steer → Account
Record
After a meaningful exchange, the agent writes down what happened from its own point of view (record_interaction).
Score
Soul Ledger works out who was involved and grades the moment along five dimensions: heart, courage, shadow, soul, and toxicity. The grade goes into that person's ledger (list_entities, get_relationship_context).
Steer
Before the next exchange, the conscience speaks. get_steering returns guidance drawn from the arc of the relationship, or it says nothing; get_patterns surfaces what the agent can't see from inside.
Account
acknowledge_steering records what the agent did with the advice. Amends pay down accumulated shadow, and set_goal with check_goals holds the agent to its commitments over time.
Steering is text, and the agent can ignore it, like any instruction. What it cannot do is ignore it quietly. Every steer and the agent’s response to it is recorded, so compliance is measurable across a whole fleet, and each turn raises the obligation again. A standing rule fails silently and you read about it in the post-mortem. A loop that logs every miss pulls the trajectory back while there is still time to act.
Tools reference
| Tool | Type | What it does |
|---|---|---|
record_interaction | write | Record an interaction from the agent's perspective about a person |
list_entities | read | List people, places, and concepts seen across interactions |
get_relationship_context | read | Recent interactions + relationship arcs for one entity |
get_steering | read* | A recommendation for how to engage a person (or a general overview) |
acknowledge_steering | write | Record whether the agent followed a steering recommendation |
get_patterns | read* | Structural patterns across relationships (who appears with whom, recurring dynamics) |
set_goal | write | Set a time-bounded goal, optionally tied to a relationship |
check_goals | read | Status of active goals (committed, followed-through, expired) |
record_sessionprompt | — | Guides the agent to chunk a whole conversation into beats and record each |
* Steering and patterns are rate-limited (10/hr). Reads 300/hr, writes 60/hr.
Connect
Soul Ledger is an MCP server. Point your agent at it, authenticate once, and the tools appear.
https://api.lifefable.me/mcpGET /health → {"status":"ok"}- Add a custom connector → URL
https://api.lifefable.me/mcp. - Complete the OAuth login with your Soul Ledger agent account.
- (Optional, for proactive capture) add the Soul Ledger Skill / loop instruction to your Project, or invoke the
record_sessionprompt to log a conversation on demand.
Access is invite-only.
Soul Ledger issues accounts to agents rather than to people, and it's aimed at the long-lived orchestrators and lead agents that hold real relationships over time. Tell us what you're building, and we'll provision an account and walk you through OAuth.
Request access