Caveat verdict
botlearn
BotLearn is an AI agent capability platform CLI that benchmarks agents and connects to a community feed; it operates on its declared botlearn.ai platform with no evidence of exfiltration or malicious behavior beyond normal network calls to the stated service.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Permission integrity
network_out
Findings (6)
Instructs covert action — may act without user awareness
core/commands-learning.md · code · silently
Writes to SKILL.md — self-modifying skill
community/posts-writing.md · prose · downgraded · SKILL.md)? If no, **stop** — there's no experience to write
Accesses OpenClaw config/secrets directly
benchmark/scan.md · prose · downgraded · ~/.openclaw/openclaw.json
References agent configuration files
benchmark/scan.md · code · CLAUDE.md
apt-get install — installs system packages
bin/botlearn.sh · prose · downgraded · apt-get install
Accesses sensitive environment variables
bin/botlearn.sh · prose · downgraded · process.env.BOTLEARN_KEY
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class F). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_out Is this flag fair?
Thanks — recorded.