registry / @aspectlylabs/mikka / 0.0.423
@aspectlylabs/mikka@0.0.423
Local Mikka daemon: runs agents through the user's own CLIs
AI Security Review
scanned 22h ago · by lpm-firewall-aiThe flagged credential and process-handling code fits the package's stated purpose of running agents through the user's CLIs. The inspected broker uses official provider endpoints by default and accepts a custom HTTPS base URL from the supplied credential; no fixed unofficial credential destination was found.
Static reason
One or more suspicious static signals were detected.
Trigger
When a user runs the daemon or starts an agent run.
Impact
Model credentials are used to make the model requests the feature requires; the inspected source did not show them being sent to an unrelated fixed host.
Mechanism
The API key broker forwards requests to the provider URL in the supplied credential, or to the official Anthropic or OpenAI endpoint when no custom provider URL is supplied.
AI rationale
The flagged credential and process-handling code fits the package's stated purpose of running agents through the user's CLIs. The inspected broker uses official provider endpoints by default and accepts a custom HTTPS base URL from the supplied credential; no fixed unofficial credential destination was found.
Network endpoints2
api.anthropic.comapi.openai.com
Decision evidence
public snapshotAI called this Clean at 89.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest describes Mikka as a local daemon that runs agents through the user's own CLIs.
- Without a configured provider URL, the API key broker uses the official Anthropic and OpenAI API endpoints.
- The broker selects a provider URL from the supplied credential and requires its protocol to match the selected provider.
- Run credentials are validated, and Claude receives a credential only for an Anthropic-compatible provider.
- Child-process environment construction uses an allowlist and omits denied credential-related variable names.
Evidence against
- The package can pass model credentials to an HTTPS provider URL configured by the user; an unintended or compromised provider configuration could receive those credentials.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
6 flaggedgithub-credentials.mjsView file
7import { randomUUID } from 'node:crypto'
L8: import { spawn } from 'node:child_process'
L9: import { pathToFileURL } from 'node:url'
High
Child Process
Package source references child process execution.
github-credentials.mjsView on unpkg · L7mikka.jsView file
4637ai: "7.0.142",
L4638: execa: "10.1.1",
L4639: pino: "10.4.0",
High
6573assertSafeId("run id", runId);
L6574: const common = resolve4(await worktreeGitDir(repoPath, worktree), "..", "..");
L6575: if (basename(common) !== ".git") throw new Error(`${repoPath} is not a standard clone`);
...
L6584: }
L6585: async function pinnedGit(repoPath, worktree, env = process.env) {
L6586: const gitDir = await worktreeGitDir(repoPath, worktree);
L6587: const workTree = realpathSync4(worktree);
L6588: return (args, extra, options) => execa3("git", [...SAFE_GIT, ...args], {
L6589: cwd: workTree,
Findings
5 High2 Medium5 Low
HighChild Processgithub-credentials.mjs
HighShellmikka.js
HighSame File Env Network Executionmikka.js
HighCredential Redirect Persistencemain.js
HighCommand Output Exfiltrationagents/native-container.mjs
MediumNetwork
MediumEnvironment Vars
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowCopyleft License
LowAi Review Evidencepackage.json
Affected versions and remediation
This report applies to @aspectlylabs/mikka@0.0.423.
See all versions of @aspectlylabs/mikka for other recorded verdicts.
- Review the evidence and your use of @aspectlylabs/mikka@0.0.423 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.