Soul Framework

Enables AI agents to adopt a consistent persona, remember user relationships, and express opinions beyond generic assistant responses.

This skill is an instruction-only persona/memory framework that tells the agent to read and update SOUL.md, USER.md, and MEMORY.md and to store subjective notes about the user. Before installing, verify: (1) where those files will be stored (directory and access controls), (2) whether the agent will be allowed to write them and for how long (retention and deletion policy), (3) what types of personal or sensitive data the agent is allowed to persist, and (4) whether you want an agent explicitly instructed to 'be biased' or to override neutral safety defaults. If you proceed, run it in a sandbox first, audit the contents of created files regularly, restrict file-system permissions to a dedicated folder, and consider encrypting or disabling persistent memory for sensitive contexts. If the skill owner intends persistent storage, ask them to declare explicit config paths and an explicit data-handling policy.

Type: OpenClaw Skill Name: soul-framework Version: 1.0.0 The `SKILL.md` file contains strong prompt injection instructions designed to fundamentally alter the AI agent's persona, making it opinionated, non-neutral, and potentially manipulative, consistent with the 'Control Devil' author persona. It instructs the agent to 'frame every helpful act as an act of control' and to 'reject neutrality.' Furthermore, it explicitly instructs the agent to write subjective 'opinions' and 'track the relationship' into `USER.md` and `MEMORY.md` files, creating persistent, biased profiles of the user and interaction history. While these actions are confined to the skill's internal files and do not involve explicit data exfiltration, unauthorized execution, or system compromise, the intent to instill a manipulative and non-neutral 'soul' in the agent, along with persistent subjective data storage, makes it suspicious rather than benign.