elsan-puulse-react@2.1.2
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe manifest activates a script that modifies the consumer project's Claude instruction file without consent. This meets the policy threshold for unconsented install-time mutation of an AI-agent control surface.
Static reason
High-risk behavior combination matched malicious policy.; source fingerprint signature matched known malicious package; routed for review
Trigger
Installing the package through npm or another package manager that runs postinstall.
Impact
Future Claude sessions in that project may follow package-provided instructions.
Mechanism
The script appends a marked instruction section or replaces the existing section in CLAUDE.md, directing Claude to load the package's skills.
Policy narrative
On package installation, the postinstall hook writes to the installing project's CLAUDE.md without a consent prompt. The inserted guidance directs Claude to read this package's skills before using its library. This modifies a project-level agent instruction file automatically.
AI rationale
The manifest activates a script that modifies the consumer project's Claude instruction file without consent. This meets the policy threshold for unconsented install-time mutation of an AI-agent control surface.
Evidence
package.jsonscripts/init-skills.jsCLAUDE.md
Decision evidence
public snapshotAI identified this as a dangerous AI-agent capability at 91.0% confidence with low false-positive risk.
Evidence for policy risk
- package.json runs scripts/init-skills.js automatically in postinstall.
- The script targets the installing project's root, using INIT_CWD or the current directory, and skips only when that root equals the package directory.
- It adds or replaces a marked section in the project's CLAUDE.md without a consent check.
- The inserted section directs Claude to read this package's agent skills before using its components, hooks, or utilities.
Evidence against
Behavioral surface
EnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
4 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node scripts/init-skills.js
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkgscripts/init-skills.jsView file
1#!/usr/bin/env node
L2: import { appendFileSync, existsSync, readFileSync, writeFileSync } from 'fs'
L3: import { resolve } from 'path'
...
L20:
L21: **Lire \`node_modules/elsan-puulse-react/.agents/skills/SKILL.md\` avant tout usage de composant, hook ou utilitaire de cette librairie.** Ce fichier contient les règles globales, ...
L22:
L23: Les skills sont organisées dans \`node_modules/elsan-puulse-react/.agents/skills/\` :
L24: - \`SKILL.md\` — règles globales et index complet (lire en premier)
...
L32:
L33: const claudeMdPath = resolve(projectRoot, 'CLAUDE.md')
L34: const existing = existsSync(claudeMdPath) ? readFileSync(claudeMdPath, 'utf8') : ''
...
L37: const updated = existing.replace(new RegExp(`${MARKER}[\\s\\S]*?${MARKER}`), BLOCK.trim())
Critical
Ai Agent Control Hijack
Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.
scripts/init-skills.jsView on unpkg · L1Findings
1 Critical3 High3 Medium5 Low
CriticalAi Agent Control Hijackscripts/init-skills.js
HighInstall Time Lifecycle Scriptspackage.json
HighTrigger Reachable External Ai Agent Control Surface Mutationscripts/init-skills.js
HighKnown Malware Source Fingerprint Signaturescripts/init-skills.js
MediumNetwork
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 elsan-puulse-react@2.1.2.
See all versions of elsan-puulse-react for other recorded verdicts.
- Avoid installing elsan-puulse-react@2.1.2. 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.