@mnemom/mnemom@0.17.10
Transparent AI agent tracing
AI Security Review
scanned 12d ago · by lpm-firewall-aiThe manifest has no install-time execution hook, and inspected flagged behavior is consistent with explicit CLI features, user configuration, and guarded self-update. The static signals do not establish malicious behavior.
Decision evidence
public snapshot- package.json has no automatic install hook; prepublishOnly only runs the build.
- The Claude settings access in dist/commands/agent.js reads configuration to resolve the selected model; the launcher wraps a status command only for Claude Code and preserves the user's existing statusLine.
- dist/lib/statusline.js reads Claude settings, wraps only a configured command statusLine, and can be disabled or bypassed when custom settings are supplied.
- The minimum-version updater only proceeds in auto mode for eligible installs and invokes npm with argument arrays, not a shell.
- OAuth code obtains metadata from the configured API base; the RC proxy documents a gateway-oriented Anthropic proxy flow.
- The agent goal command is an explicit CLI control for gateway-inferred session goals.
Source & flagged code
6 flagged · loading sourcePackage source references child process execution.
dist/lib/update/min-version.jsView on unpkg · L6Package source references shell execution.
dist/lib/oauth.js#virtual:normalized:round1View on unpkg · L310Findings
Affected versions and remediation
This report applies to @mnemom/mnemom@0.17.10.
See all versions of @mnemom/mnemom for other recorded verdicts.
- Review the evidence and your use of @mnemom/mnemom@0.17.10 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.