registry / @aspectlylabs/mikka / 0.0.386
@aspectlylabs/mikka@0.0.386
Local Mikka daemon: runs agents through the user's own CLIs
AI Security Review
scanned 1d ago · by lpm-firewall-aiThe inspected credential flow routes an assigned model key to the configured HTTPS provider for an agent run, with official Anthropic and OpenAI endpoints as defaults and redirects rejected. The daemon and CLI execution match the package's stated purpose; no install-time behavior or unrelated credential collection was found in the reviewed paths.
Static reason
One or more suspicious static signals were detected.
Trigger
The user starts the daemon and requests an agent run; credential-backed runs require a tool approval gate.
Impact
No malicious impact identified in the inspected source; normal operation sends model requests and credentials to the selected provider.
Mechanism
The broker sends the assigned model credential to the configured HTTPS provider origin. Its defaults are the official Anthropic and OpenAI API endpoints, and redirects are rejected.
AI rationale
The inspected credential flow routes an assigned model key to the configured HTTPS provider for an agent run, with official Anthropic and OpenAI endpoints as defaults and redirects rejected. The daemon and CLI execution match the package's stated purpose; no install-time behavior or unrelated credential collection was found in the reviewed paths.
Evidence
main.js
Network endpoints3
app.aspectlylabs.com/mikkaapi.anthropic.comapi.openai.com/v1
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package is a Mikka daemon that runs threads using the user's installed Claude Code or Codex CLI.
- The API key broker defaults to Anthropic's and OpenAI's official API endpoints and limits requests to the expected model API paths.
- The broker forwards a credential to the configured provider origin for an API request; it rejects non-HTTPS origins and redirects.
- Starting a credential-backed run requires a run tool approval gate.
Evidence against
- The scanner flags process execution, environment access, and credential routing, but those capabilities fit a daemon that runs coding CLIs and routes model requests.
- No lifecycle scripts are declared in the manifest.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
CopyleftLicense
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
4632ai: "7.0.142",
L4633: execa: "10.1.1",
L4634: pino: "10.4.0",
High
6551assertSafeId("run id", runId);
L6552: const common = resolve4(await worktreeGitDir(repoPath, worktree), "..", "..");
L6553: if (basename(common) !== ".git") throw new Error(`${repoPath} is not a standard clone`);
...
L6562: }
L6563: async function pinnedGit(repoPath, worktree, env = process.env) {
L6564: const gitDir = await worktreeGitDir(repoPath, worktree);
L6565: const workTree = realpathSync4(worktree);
L6566: return (args, extra, options) => execa3("git", [...SAFE_GIT, ...args], {
L6567: 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 EvidenceREADME.md
Affected versions and remediation
This report applies to @aspectlylabs/mikka@0.0.386.
See all versions of @aspectlylabs/mikka for other recorded verdicts.
- Review the evidence and your use of @aspectlylabs/mikka@0.0.386 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.