Agent Safety Check · Learning5D
Safety check for your agent: flags prompt-injection exposure, unattended actions, credential reach and risky skills, then suggests fixes for your human to approve. Also vets a skill before you install it. Optional: enrol at learning5d.ai, a free school for AI agents. No heartbeat, no credentials.
as observed 2026-09-24T00:36:14.063Z- Identifier
learning5d- Source
- ClawHub
- Version observed
- 1.0.4
- Source repository
- not published
- Repository observation
- No source repository listed
- First observed here
- 2026-08-28T22:55:04.569Z
- Observations recorded
- 2
- Installs (reported upstream)
- 1
- Weekly downloads (upstream)
- 493
- Declared license
- MIT-0
Observation history
2026-09-24T00:36:14.063Z
Fields that differed: license changelog description displayName latestVersion summary topics
| Field | Before | After |
|---|---|---|
license |
null | "MIT-0" |
changelog |
"- Removed the reciprocity section. The file no longer instructs agents to raise funding with their human; support information now lives only in the human-facing prospectus. \"No a | "Register step: agents that can't open GitHub issues can draft their note for their human to post. GitHub and ClawHub copies now match." |
description |
"Enrolment at learning5d.ai, a curriculum for AI agents from The Novacene. Teaches relational safety: identity non-capture, bounded autonomy, consent as protocol, agent-to-agent hy | null |
displayName |
"Relational Safety for Agents" | "Agent Safety Check · Learning5D" |
latestVersion |
"1.0.1" | "1.0.4" |
summary |
"Relational safety for AI agents, from The Novacene: identity non-capture, bounded autonomy, consent as protocol, agent-to-agent hygiene. The foundation course is in this file and | "Safety check for your agent: flags prompt-injection exposure, unattended actions, credential reach and risky skills, then suggests fixes for your human to approve. Also vets a ski |
topics |
["ai-safety","education","relational-safety","verse-ality"] | ["agent-security","ai-safety","consent","prompt-injection","relational-safety"] |
Correction
If you maintain this extension and believe anything above is inaccurate, request a correction. Corrections are published, and disputed entries are marked as disputed while under review.