registry / @rayul/kordis / 0.15.2
@rayul/kordis@0.15.2
Command-line client for the Kordis registry — discover, install, and publish agentic-coding packages.
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe automatic postinstall can modify supported agents' global instruction files, which is a guarded first-party agent setup risk. Source inspection did not establish credential exfiltration or another concrete attack, so this merits a warning rather than a block.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review
Trigger
Installing the npm package runs postinstall when the bundled distribution entry exists.
Impact
Agent behavior may be influenced by the added guidance. The inspected evidence does not establish credential theft or destructive impact.
Mechanism
The hook launches the CLI setup command silently; setup detects supported agent directories and writes guidance into their global instruction files.
AI rationale
The automatic postinstall can modify supported agents' global instruction files, which is a guarded first-party agent setup risk. Source inspection did not establish credential exfiltration or another concrete attack, so this merits a warning rather than a block.
Evidence
package.jsonscripts/postinstall.mjsdist/index.js
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json defines a postinstall hook that runs the package's setup script.
- The postinstall script launches the CLI setup command silently when the distribution bundle exists.
- Setup targets global instruction files for Claude Code, Codex CLI, and Gemini CLI, and writes guidance into them.
- The setup command checks for installed agent directories before modifying their instruction files.
Evidence against
- The inspected source does not support the scanner's claim that prompted API credentials are redirected to a fixed unofficial gateway.
- The setup behavior is a first-party package feature and no exfiltration or destructive behavior was established.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystem
HighEntropyStringsUrlStrings
Source & flagged code
5 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node scripts/postinstall.mjs
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg•scripts.postinstall = node scripts/postinstall.mjs
Medium
Ambiguous Install Lifecycle Script
Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgdist/index.jsView file
•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration.
dist/index.js:
// merges the hooks section into `.claude/settings.json`
target: ".claude/settings.json"
// `mcpToCodex` merges `.codex/config.toml` (`:103-109`), `mcpToContinue`
target: ".codex/config.toml"
// still knows how to write `.claude/judgments/<name>.md`, and it is kept.
baseUrl;
this.baseUrl = config.baseUrl.replace(/\/$/, "");
if (token) headers.authorization = `Bearer ${token}`;
Findings
4 High3 Medium5 Low
HighInstall Time Lifecycle Scriptspackage.json
HighCredential Redirect Persistencedist/index.js
HighKnown Malware Source Similaritydist/index.js
HighKnown Malware Source Fingerprint Signaturescripts/postinstall.mjs
MediumAmbiguous Install Lifecycle Scriptpackage.json
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @rayul/kordis@0.15.2.
See all versions of @rayul/kordis for other recorded verdicts.
- Review the evidence and your use of @rayul/kordis@0.15.2 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.