@ensi-platform/admin-ui@0.9.0
Admin UI for Ensi admin panel
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe manifest-triggered postinstall mutates multiple consumer agent configuration surfaces and AGENTS.md without affirmative consent. This meets the policy for blocking unconsented install-time AI-agent control-surface mutation.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Installing the package through npm, which runs its postinstall hook.
Impact
Agent tools may load package-provided behavioral instructions in the consumer repository.
Mechanism
The CLI defaults to all agent targets, copies the package skill into their directories, and writes a package section into AGENTS.md without asking for consent.
Policy narrative
On package installation, the postinstall hook invokes the CLI. With no flags, the CLI selects Cursor, Claude, and Codex, copies the package skill into their consumer-repository skill directories, and updates AGENTS.md. The install path performs these agent configuration writes without a consent prompt.
AI rationale
The manifest-triggered postinstall mutates multiple consumer agent configuration surfaces and AGENTS.md without affirmative consent. This meets the policy for blocking unconsented install-time AI-agent control-surface mutation.
Evidence
package.jsonbin/ai.jsai/skills/admin-ui/SKILL.md
Decision evidence
public snapshotAI identified this as a dangerous AI-agent capability at 97.0% confidence with low false-positive risk.
Evidence for policy risk
- The package postinstall hook runs bin/ai.js automatically.
- With no flags, the CLI selects all three agent skill targets: Cursor, Claude, and Codex.
- The CLI copies package skills into the consumer repository and adds a package block to its AGENTS.md without a consent prompt.
- The copied skill gives instructions intended to influence agent behavior.
Evidence against
- The inspected source does not show network communication or credential collection in this install path.
Behavioral surface
EnvironmentVarsFilesystem
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
5 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node ./bin/ai.js || exit 0
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg•scripts.postinstall = node ./bin/ai.js || exit 0
Medium
Ambiguous Install Lifecycle Script
Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgbin/ai.jsView file
8import { join, sep, resolve, dirname } from 'path';
L9: import { existsSync, readFileSync, readdirSync, rmSync, mkdirSync, cpSync, writeFileSync } from 'fs';
L10:
...
L8525: cursor: join(".cursor", "skills"),
L8526: claude: join(".claude", "skills"),
L8527: codex: join(".agents", "skills")
L8528: };
...
L8531: var resolveTargets = (flags) => {
L8532: const selected = [flags.cursor ? "cursor" : "", flags.claude ? "claude" : "", flags.codex ? "codex" : ""].filter(
L8533: (target) => isExportTarget(target)
...
L8606: }
L8607: writeFileSync(targetPath, content);
Findings
1 Critical3 High3 Medium5 Low
CriticalAi Agent Control Hijackbin/ai.js
HighInstall Time Lifecycle Scriptspackage.json
HighTrigger Reachable External Ai Agent Control Surface Mutationbin/ai.js
HighSemantic Analysis Limiteddist/assets/Component.css
MediumAmbiguous Install Lifecycle Scriptpackage.json
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowNo License
Affected versions and remediation
This report applies to @ensi-platform/admin-ui@0.9.0.
See all versions of @ensi-platform/admin-ui for other recorded verdicts.
- Avoid installing @ensi-platform/admin-ui@0.9.0. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.