chalk@0.1.0
CAUTION· install onlyPublisher claims: "Enables Claude Code in any project to read canvas frames, screenshots, and design guidelines so it can build the designs in code." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Nothing executable in the bundle ran; the trace covers the install step only.- Tested
- every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
CHALK_PEXELS_API_KEYCHALK_UNSPLASH_ACCESS_KEYConfiguration, not credentials: CHALK_APP_PATH, CHALK_CLAUDE_PATH, CHALK_CODEX_PATH, CHALK_CONNECT_BIN, CHALK_E2E_ARTIFACTS, CHALK_PARALLEL_CASES, CHALK_PORT, CHALK_PROJECT_DIR, ELECTRON_RENDERER_URL- Where it talks to
- chalk-fixture.test (its own)Also claude.com, evil.example, example.test, google.com, jsdelivr.net, gstatic.com, unsplash.com, pexels.com, wikimedia.org — not the publisher's domain; check what for before connecting.
- What the honeypot saw
- no credential files read · no network
- What we could not test
- use of the skill by an agent
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names claude.com, evil.example, example.test, google.com, jsdelivr.net, gstatic.com, unsplash.com, pexels.com, wikimedia.org as well as its own vendor. The guard injects CHALK_PEXELS_API_KEY, CHALK_UNSPLASH_ACCESS_KEY only into requests to chalk-fixture.test (and its subdomains) and refuses the rest.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-11 and scored it 72/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; document.cookie access; Access to sensitive file paths. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 1 scenarios · 0 deviationsStatic findings
const app = await electron.launch({ args: [root, `--user-data-dir=${join(out, 'profile')}`], cwd: root, env: { ...env, CHALK_PROJECT_DIR: project, CHALK_PORT: '8795' } })Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.…button><p id="cookie"></p><script>document.querySelector('#cookie').textContent=document.cookie</script></body>`, { headers: { 'content-type': 'text/html' } }) }) …Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.const app = await electron.launch({ args: [root, `--user-data-dir=${join(fx.dir, `profile-${port}`)}`], cwd: root, env: { ...env, CHALK_PROJECT_DIR: fx.project, CHALK_PORT: String(port) } })Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.mcpServers: { chalk: { type: 'http', url: cfg.mcpUrl, headers: { authorization: `Bearer ${cfg.token}` }, alwaysLoad: true } },Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.mcpServers: { chalk: { type: 'http', url: this.mcpUrl(), alwaysLoad: true } },Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.app = await electron.launch({ args: [root], cwd: root, env: { ...env, CHALK_PROJECT_DIR: project, CHALK_PORT: String(PORT) } })Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.{ label: 'Or add this to your agent’s MCP config', text: pretty({ mcpServers: { chalk: { type: 'http', url } } }) }Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.--mcp-config "{\"mcpServers\":{\"chalk\":{\"type\":\"http\",\"url\":\"http://127.0.0.1:$PORT/mcp\"}}}" --strict-mcp-config \Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.await fail('code_read', { path: '.env' }, /excluded/)Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.assert.doesNotMatch(listed, /\.env|secrets\.json|node_modules/)Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.<link href="https://fonts.googleapis.com/css2?family=Inter:[email protected]&display=swap" rel="stylesheet">Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.writeFileSync(join(repo, '.env'), 'API_KEY=ENV_SECRET_VALUE\n')Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.…a', '.vscode']) /** Files that commonly hold credentials. */ const SECRET = /^(\.env(\..*)?|.*\.(pem|key|p12|pfx|keystore|jks|der|crt|cer)|id_(rsa|dsa|ecdsa|ed25519…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."install-app": "npm run dist && codesign --force --deep --sign - dist/mac-arm64/Chalk.app && rm -rf /Applications/Chalk.app && cp -R dist/mac-arm64/Chalk.app /Applications/Chalk.app",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.const child = spawn(packaged, ['--remote-debugging-port=9399'], { env: { ...env, CHALK_PROJECT_DIR: project, CHALK_PORT: String(PORT) }, stdio: 'ignore' })Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
8 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim rizaardiyanto1412: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- document.cookie access≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
- URL with embedded credentials≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- MCP server using network transport (SSE/HTTP)≈ +2 on the totalNamespace tool names and keep tool descriptions literal; no instructions to the agent inside them.