sneakoscope@10.2.0
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-aiThis is a guarded agent-extension lifecycle risk, not a confirmed malicious install chain. Automatic code execution and incomplete source coverage justify a warning rather than a clean verdict.
Decision evidence
public snapshot- An npm postinstall hook automatically runs the SKS command.
- The postinstall handler writes a build stamp inside the installed package.
- Opt-in postinstall bootstrap can target a project or global SKS location and forces bootstrap.
- The package contains broad Codex and agent-management command surfaces.
- Release and fixture helpers can launch shell commands, increasing lifecycle review risk.
- A bundled native PDF resource limits complete source-level coverage.
- The default postinstall branch returns before external project, home, Codex, or global-state changes.
- Broader bootstrap requires SKS_POSTINSTALL_BOOTSTRAP=1.
- No runtime dependency on sneakoscope itself is declared.
- No credential collection or network exfiltration was confirmed in the inspected postinstall path.
Source & flagged code
14 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@10.2.0.
See all versions of sneakoscope for other recorded verdicts.
- Review the evidence and your use of sneakoscope@10.2.0 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.