Caveat verdict
agent-recall
The agent-recall skill is designed for persistent memory needs, operating locally without cloud or telemetry involvement, indicating a focus on user privacy.
⚠ 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.
Findings (8)
Recursive delete from root or home — destructive command
commands/arstatus.md · prose · downgraded · rm -rf ~
Instructs covert action — may act without user awareness
packages/cli/README.md · code · silently
Pipe to python — executes piped content as Python code
docs/sdk-design.md · prose · downgraded · | Python
References agent memory files
commands/arstart.md · code · MEMORY.md
Dynamic import() — loads module at runtime
packages/cli/src/index.ts · prose · downgraded · import("
Uses exec() — may execute shell commands
packages/cli/src/utils/transcript-reader.ts · prose · downgraded · exec(
References child_process — can spawn system processes
packages/core/src/storage/project.ts · prose · downgraded · child_process
Popular HTTP library — network access
commands/arsave.md · prose · downgraded · got
Permissions & capabilities
No declared permissions — minimal attack surface.
network_in Is this flag fair?
Thanks — recorded.