registry / @makerx/yoke / 0.6.0
@makerx/yoke@0.6.0
Lean, user-invoked Claude Code skills for MakerX AI-assisted delivery, plus the CLI they call.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe package is a user-invoked CLI with no install hook. Its external process, provider request, GitHub posting, and skill-writing behavior is tied to review, optional posting, or an explicit sync command; the inspected source showed no attack behavior.
Static reason
One or more suspicious static signals were detected.
Trigger
User runs a CLI command; provider probes occur during model or review operations, and GitHub posting requires `--post`.
Impact
No unauthorized credential transfer, install-time activity, or unrequested account action was identified.
Mechanism
The CLI invokes Claude or Codex for reviews, sends configured provider credentials to the configured provider endpoint, and writes bundled skills when `yoke sync` is run.
AI rationale
The package is a user-invoked CLI with no install hook. Its external process, provider request, GitHub posting, and skill-writing behavior is tied to review, optional posting, or an explicit sync command; the inspected source showed no attack behavior.
Evidence
.claude/skills/<skill>/SKILL.md~/.claude/skills/<skill>/SKILL.md
Network endpoints2
User-configured provider base URLgithub.com
Decision evidence
public snapshotAI called this Clean at 92.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest exposes dist/cli.js as a command-line executable and has no install, preinstall, or postinstall hook.
- The package describes itself as user-invoked Claude Code skills and a CLI.
- Provider probes send API keys to endpoints derived from the provider base URL configured by the user.
- Skill writes happen through the explicit `yoke sync` command, with user-level writes requiring `--user`.
- GitHub posting is an optional `--post` review action.
Evidence against
- The CLI runs external Claude or Codex processes and handles their output as part of its review function.
- The CLI can write skill instructions under `.claude/skills`, but this behavior is explicitly invoked with `yoke sync`.
Behavioral surface
ChildProcessEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
4 flaggeddist/cli.jsView file
282// src/features/pr/raise.ts
L283: import { execFileSync } from "child_process";
L284:
High
1120async function claudeSignedIn() {
L1121: const { ok, stdout } = await exec("claude", ["auth", "status", "--json"]);
L1122: if (!ok) return false;
...
L1131: try {
L1132: const response = await fetch(url, {
L1133: method: "POST",
L1134: headers: { "Content-Type": "application/json", ...headers },
L1135: body: JSON.stringify(body),
L1136: signal: AbortSignal.timeout(probeTimeout)
High
Command Output Exfiltration
Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.
dist/cli.jsView on unpkg · L1120•Runtime or CLI code links a write operation to an explicit external AI-agent control path:
pr raise --help\` and follow it exactly. do not use \`gh pr create\`.
if \`yoke\` is not found, run \`npm i -g @makerx/yoke\` and retry.
`
};
var reviewskill = {
name: "review",
description: "adversarially review a pr with claude and codex, then combine their findings.",
disablemodelinvocation: true,
body: `
run \`yoke review $arguments\` in the foreground, not as a background task, with the longest timeout your shell tool allows, so the user sees each reviewer's progress as it streams; two reviewers and a synthesis pass take several minutes. then relay the findings it printed. do not add findings of your own or re-review the code. each reviewer's tool calls
Findings
5 High2 Medium5 Low
HighChild Processdist/cli.js
HighShell
HighCommand Output Exfiltrationdist/cli.js
HighExternal Ai Agent Control Surface Mutationdist/cli.js
HighTrigger Reachable Command Output Exfiltrationdist/cli.js
MediumNetwork
MediumEnvironment Vars
LowNon Install Lifecycle Scripts
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @makerx/yoke@0.6.0.
See all versions of @makerx/yoke for other recorded verdicts.
- Review the evidence and your use of @makerx/yoke@0.6.0 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.