registry / ocean-brain / 0.15.1
ocean-brain@0.15.1
⚠ Under review<!-- npm package README for the public ocean-brain CLI.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiInspected source did not independently support a warning or block, and scanner labels alone are insufficient evidence. Manual review is required because the mandated nine distinct clean citations exceed the maximum of eight.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review
Trigger
npm postinstall runs scripts/postinstall-prisma.mjs.
Impact
No malicious effect was confirmed; the required clean citation coverage cannot fit the response limit.
Mechanism
The script resolves the bundled Prisma CLI and invokes its generate command with a schema path; inspected flagged primitives did not establish malicious execution.
AI rationale
Inspected source did not independently support a warning or block, and scanner labels alone are insufficient evidence. Manual review is required because the mandated nine distinct clean citations exceed the maximum of eight.
Decision evidence
public snapshotAI called this Manual Review at 98.0% confidence as Unknown with high false-positive risk.
AI review observations
- package.json activates a postinstall script that runs the bundled Prisma CLI with the generate command.
- The inspected Prisma preinstall entry checks the Node.js version.
- The SQLite search implementation loads the sqlite-vec extension for its search database.
- No concrete attack was established from inspected source; clean publication requires nine distinct citations, exceeding the eight-citation limit.
Evidence against
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedPossibleObfuscationTelemetryUrlStrings
Source & flagged code
14 flaggedpackage.jsonView file
•scripts.postinstall = node ./scripts/postinstall-prisma.mjs
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkgserver/dist/features/search/sqlite-vector-index.jsView file
76const initializeSearchDatabase = async (database) => {
L77: await exec(
L78: database,
High
Child Process
Package source references child process execution.
server/dist/features/search/sqlite-vector-index.jsView on unpkg · L76server/client/dist/assets/NotePropertiesPanel-DI1AICh8.jsView file
Findings
2 Critical9 High5 Medium6 Low
CriticalRemote Asset Decode Executenode_modules/prisma/build/index.js
CriticalTrojan Source Unicodenode_modules/confbox/dist/_chunks/libs/json5.mjs
HighInstall Time Lifecycle Scriptspackage.json
HighChild Processserver/dist/features/search/sqlite-vector-index.js
HighEvalserver/client/dist/assets/NotePropertiesPanel-DI1AICh8.js
HighSame File Env Network Executionnode_modules/@prisma/client/runtime/binary.js
HighCommand Output Exfiltrationnode_modules/@prisma/client/runtime/binary.js
HighObfuscated Payload Loadernode_modules/prisma/build/index.js
HighCross File Remote Execution Contextnode_modules/@prisma/client/generator-build/index.js
HighSemantic Analysis Limited
Affected versions and remediation
This report applies to ocean-brain@0.15.1.
See all versions of ocean-brain for other recorded verdicts.
- Review the evidence and your use of ocean-brain@0.15.1 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.