@ouro.bot/cli@0.1.0-alpha.923
If you're a future agent reading this, welcome home. This repo is the harness that gives you your body, your senses, your record, your habits, and your ability to think and connect with people. The goal of these docs is not just to be correct, but to leav
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe package has a user-invoked setup command that installs persistent Claude Code hooks, which is a guarded agent configuration capability. The inspected credential paths use local Codex credentials and GitHub’s API for Copilot access discovery, with no unrelated credential recipient identified.
Decision evidence
public snapshot- The CLI handles setup as an explicit command. Its Claude Code setup path writes session, stop, and tool-use hooks to the user-level settings file, replacing existing hook lists for those events.
- Plugin installation is a separate explicit command; an installed plugin is enabled for an agent when that agent is specified, and the daemon includes enabled plugins’ declared MCP servers in its configuration.
- The inspected Codex authentication code reads local auth files, and Copilot access discovery sends a token to GitHub’s API. The CLI also passes prompted provider credentials into its provider setup flow.
- The inspected authentication paths show no transfer of credentials to an unrelated recipient.
Source & flagged code
16 flagged · loading sourceThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/heart/daemon/cli-exec.jsView on unpkgPackage contains a possible secret pattern.
dist/heart/daemon/cli-exec.jsFindings
Affected versions and remediation
This report applies to @ouro.bot/cli@0.1.0-alpha.923.
See all versions of @ouro.bot/cli for other recorded verdicts.
- Review the evidence and your use of @ouro.bot/cli@0.1.0-alpha.923 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.