oh-my-claude-sisyphus@4.15.10
Multi-agent orchestration system for Claude Code - Inspired by oh-my-opencode
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe scanner's claimed default credential exfiltration is not supported by the source: network calls are configured notification/update features and the daemon uses an environment allowlist. The remaining extension lifecycle and remote-control capability warrants a warning, not a block.
Decision evidence
public snapshot- Runtime update runs global npm install and reconciles Claude artifacts.
- CLI launch creates a scoped .omc-launch Claude runtime and links credentials.
- Reply listener can inject configured Discord/Telegram/Slack replies into tmux.
- Silent auto-update can apply updates after explicit configuration.
- package.json has no preinstall/install/postinstall hook.
- Release checks target the declared GitHub repository.
- Silent auto-update defaults disabled without configuration.
- Notification endpoints and tokens are configuration-dependent; no package-owned exfil receiver found.
Source & flagged code
15 flaggedPackage source references child process execution.
dist/autoresearch/runtime.jsView on unpkg · L1Package source references a known benign dynamic code generation pattern.
Findings
Affected versions and remediation
This report applies to oh-my-claude-sisyphus@4.15.10.
See all versions of oh-my-claude-sisyphus for other recorded verdicts.
- Review the evidence and your use of oh-my-claude-sisyphus@4.15.10 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.