registry / @agent-native/skills / 0.3.41
@agent-native/skills@0.3.41
⚠ Under reviewInstall BuilderIO skills for coding agents.
AI Security Review
scanned 15d ago · by lpm-firewall-aiThe package’s standard CLI path invokes an unresolved runtime dependency, leaving delegated behavior unverified. No independent malicious behavior was established, so manual review is appropriate.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Running the CLI without setting AGENT_NATIVE_SKILLS_DIRECT to 1.
Impact
The delegated behavior cannot be verified from this snapshot, so malicious impact is unconfirmed.
Mechanism
The package dynamically imports runSkills from @agent-native/core/cli/skills; the inspected direct installer also supports selected MCP registration and interactive authentication.
AI rationale
The package’s standard CLI path invokes an unresolved runtime dependency, leaving delegated behavior unverified. No independent malicious behavior was established, so manual review is appropriate.
Evidence
package.jsondist/index.js
Network endpoints1
analytics.agent-native.com/track
Decision evidence
public snapshotAI called this Manual Review at 82.0% confidence as Unknown with medium false-positive risk.
AI review observations
- The manifest declares @agent-native/core 0.212.0, and the normal CLI path dynamically imports its runSkills implementation; that delegated behavior remains unresolved in this snapshot.
- The manifest has no npm install lifecycle hook, and the package’s executable entry invokes the CLI with user-supplied arguments.
- Agent setup is tied to selected installation options; device authentication requires an interactive flow and browser approval before credentials are written to agent config.
- The CLI sends usage events with a generated installation identifier to the package’s first-party analytics endpoint.
Evidence against
- No concrete malicious action was established in the inspected package source.
- The direct source shows interactive MCP authentication and config writes as part of the installer flow.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
Source & flagged code
5 flaggeddist/index.jsView file
1import { spawnSync } from "node:child_process";
L2: import { randomUUID } from "node:crypto";
High
•matchType = previous_version_dangerous_delta
matchedPackage = @agent-native/skills@0.2.724
matchedIdentity = npm:QGFnZW50LW5hdGl2ZS9za2lsbHM:0.2.724
similarity = 0.500
summary = stored previous version shares package body but lacks this dangerous source file
High
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/index.jsView on unpkgdist/index.js#virtual:normalized:round1View file
32--force Overwrite a different existing PR Visual Recap workflow
L33: --no-mcp Install skill files only; skip registering the app"s MCP server\n --no-connect Register MCP where possible but skip inline browser/devic...
L34: return file;
Findings
5 High2 Medium5 Low
HighChild Processdist/index.js
HighShell
HighSame File Env Network Executiondist/index.js#virtual:normalized:round1
HighSemantic Analysis Limiteddist/index.js#virtual:normalized:round1
HighPrevious Version Dangerous Deltadist/index.js
MediumNetwork
MediumEnvironment Vars
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowAi Review Evidencepackage.json
Affected versions and remediation
This report applies to @agent-native/skills@0.3.41.
See all versions of @agent-native/skills for other recorded verdicts.
- Review the evidence and your use of @agent-native/skills@0.3.41 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.