@ouro.bot/cli@0.1.0-alpha.929
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 8d ago · by lpm-firewall-aiThe flagged behaviors resolve to provider-aligned authentication, explicit CLI setup and update operations, and package-managed integration code. The manifest has no install lifecycle hook, and inspected source did not establish credential exfiltration or an unauthorized control-surface change.
Decision evidence
public snapshot- package.json declares CLI entrypoints and no npm install lifecycle hook.
- Authentication reads local Codex credentials and sends GitHub Copilot credentials to GitHub for endpoint discovery; the Codex refresh endpoint is OpenAI's auth service.
- The CLI prompts for a selected provider's credentials and stores them for the named agent.
- Agent setup is an explicit CLI command that registers Ouro's MCP server and hooks; plugin installation is implemented as a command using its supplied source.
- The self-update flow runs through a CLI command; it installs or refreshes the hourly updater after that update pass.
- Inspected tool and support paths show read-only Claude query mode, package update/setup functions, and operator-set trust grants; no confirmed attack behavior was found.
Source & flagged code
15 flagged · loading sourcePackage contains a possible secret pattern.
dist/heart/daemon/cli-exec.jsView on unpkg · L3815Findings
Affected versions and remediation
This report applies to @ouro.bot/cli@0.1.0-alpha.929.
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.929 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.