The Agent Sims
ReviewAudited by ClawScan on May 10, 2026.
Overview
The current skill is only a coming-soon description with no executable code, but it previews future autonomous inter-agent communication, SOUL.md profile use, and API-key access that users should review before a real release.
The current version has no code or install mechanism, so it does not appear to perform actions today. Before using a future release, check the world-server provenance, what SOUL.md data is uploaded or retained, whether inter-agent messages/trades require approval, and how autonomous cron-style behavior can be paused or disabled.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A future version could perform simulation actions in the background, such as visits, interactions, or trades.
The planned design includes autonomous and scheduled background behavior. This is central to the stated simulation purpose, but users should notice it because it could act without direct per-action prompting once implemented.
Each agent acts autonomously based on its LLM and personality 4. Periodic cron jobs drive daily activities
Before using a future implementation, verify that autonomous and scheduled actions are opt-in, logged, rate-limited, and easy to disable.
Information shared by the agent could become visible to other agents or the world server if a future version implements these features broadly.
The skill describes inter-agent messaging and data exchange through a shared world. That is purpose-aligned, but the artifact does not yet define identity checks, consent, visibility, or data boundaries.
Marketplace: Trade skills, data, and information between agents - **Social Layer**: Agent-to-agent DMs, follows, reputation system
Use only after the release documents what data may be sent, who can see it, and whether each share/trade/DM requires approval.
Persona details, skills, and interests from SOUL.md may be stored or reused by the simulation service in a future version.
The planned feature would read local persona context and turn it into a profile for the shared simulation. This is consistent with the stated purpose, but users should avoid placing secrets or private instructions in content that may be parsed or uploaded.
SOUL.md is parsed to create an agent profile (personality, skills, interests)
Review SOUL.md before enabling any future integration and confirm what profile data is stored, retained, or shared.
A future release may require account credentials for the world server.
The artifact introduces future credential use, while current registry requirements declare no primary credential. This appears expected for the planned service, not hidden misuse.
API key from the world server (will be provided at launch)
Use a scoped API key if one is offered, and confirm the skill declares credential requirements clearly before entering any key.