Openclaw Default Agent Backstory

Build, bootstrap, and maintain a stable OpenClaw default agent identity by interviewing the user, updating IDENTITY.md, and seeding core context files (AGENT...

This skill is internally coherent and appears safe to use, but it will read and write files at your workspace root (IDENTITY.md, AGENTS.md, MEMORY.md, etc.). Before running: 1) back up or version-control your workspace so you can review diffs; 2) avoid pasting secrets or sensitive credentials into answers during the interview; 3) confirm any automated file updates before committing changes; and 4) if you want stricter control, ask the agent to show proposed file contents and require your explicit approval before writing.

Type: OpenClaw Skill Name: openclaw-default-agent-backstory Version: 1.0.0 This skill bundle is designed to help an OpenClaw agent establish and maintain its identity and context files. The `SKILL.md` and supporting documentation (`references/bootstrap-core-files.md`, `references/question-bank.md`) clearly define instructions for the agent to read existing context files, interview the user, and then write or update specific Markdown files (`IDENTITY.md`, `AGENTS.md`, `SOUL.md`, etc.) and create a `memory/` directory. All file operations are directly aligned with the stated purpose of building and refining agent identity. There is no evidence of intent for data exfiltration, malicious execution, persistence, or prompt injection designed to subvert the agent's core function or access unrelated sensitive data. The instructions are explicit about the files to be written and the type of content, which is necessary for the skill's functionality.