sneakoscope@10.0.0
Sneakoscope Codex (`sks`) is an open-source trust layer for Codex CLI and ChatGPT Desktop. It coordinates bounded AI coding agents, records machine-verifiable evidence, preserves project memory, and blocks release claims that are not supported by current
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe risky behavior is explicitly gated, so it is not an unconsented automatic install-time takeover. Its global agent configuration changes and unpinned external execution still warrant a warning.
Decision evidence
public snapshot- The package automatically runs a postinstall command.
- An environment-variable opt-in enables postinstall bootstrap work.
- That opt-in path adds a global Codex MCP server using an unpinned latest package.
- That same path installs and removes generated global Codex skills in the user's home directory.
- Without the opt-in variable, the postinstall path returns before external setup.
- The MCP setup clears the CodeLB API key from the child environment.
- The database policy defaults to read-only behavior and blocks destructive operations.
- No self-dependency or automatic credential-exfiltration path was found in the inspected sources.
Source & flagged code
15 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage contains a possible secret pattern.
dist/core/codex/managed-proxy-env.jsView on unpkg · L43Findings
Affected versions and remediation
This report applies to sneakoscope@10.0.0.
See all versions of sneakoscope for other recorded verdicts.
- Review the evidence and your use of sneakoscope@10.0.0 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.