Back to skill

Security audit

Deplark / 速栈 — 快速部署站点,中国大陆高速访问

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because the partial scan found credential-storage access and detected environment and network capabilities without a declared tool scope; it did not report clear malicious intent.

Review the skill’s declared scope and how it uses credentials before installing, and decide whether its environment and network access fit your needs. The scan covered 66.7% of components and had partial analysis for some checks, so this verdict is based on incomplete evidence.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (21)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 110)May include surrounding context.

js
child.stdout.on('data', data => { if (output.length < 16384) output += data.toString(); });
    child.stderr.on('data', () => {}); // Native diagnostics can contain sensitive arguments.
    child.on('error', () => { clearTimeout(timer); if (!finished) { finished = true; reject(problem('LOCAL_CLIENT_UNAVAILABLE')); } });
    child.on('close', code => { clearTimeout(timer); if (!finished) { finished = true; code === 0 ? resolveResult(output) : reject(problem('KEYCHAIN_UNAVAILABLE')); } });
    child.stdin.on('error', () => {}); child.stdin.end(input || '');
  });
}

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 118)May include surrounding context.

js
const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 130)May include surrounding context.

js
// security -i reads commands on stdin. The password never appears in argv.
      await run('/usr/bin/security', ['-i'], `add-generic-password -U -a ${account} -s ${service} -w ${token}${suffix}\n`);
      // Interactive security may exit zero after a failed command; verify the write.
      const saved = await this.get(); if (saved !== token) throw problem('KEYCHAIN_SAVE_FAILED');
    },
    async delete() { try { await run('/usr/bin/security', ['delete-generic-password', '-a', account, '-s', service, ...(keychain ? [keychain] : [])]); } catch {} },
  };

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 319)May include surrounding context.

js
if(flags['--api-key'])return {...await client.connectExisting(await secretInput(),flags['--verification']),authorization_source:'existing-key'};
    if(flags['--resume'])return client.resumeBrowser({verification:flags['--verification']});
    if(!flags['--browser'])try{return {...await client.connectSaved(flags['--verification']),authorization_source:'saved-key'};}catch(error){if(!['NO_LOCAL_AUTHORIZATION','ACCOUNT_SKILL_KEY_INVALID'].includes(error.code))throw error;}
    if(process.platform!=='darwin')throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY');
    const result=await client.startBrowser({verification:flags['--verification']});
    if(!flags['--no-open']){const child=spawn('/usr/bin/open',[result.authorization_url],{stdio:'ignore'});child.on('error',()=>{});child.unref();}
    return result;

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 381)May include surrounding context.

js
if(account)return runAccountCLI(command,flags,{origin,agent,scope,manifest,name,runtime});
  const store = process.platform === 'darwin' ? keychainStore({origin, project, agent}) : {
    async get() { return process.env.DEPLARK_API_KEY || null; },
    async set(value) { if (value !== process.env.DEPLARK_API_KEY) throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY'); }, async delete() {},
  };
  const client = createSkillClient({origin, project, agent, store, apiKey: process.env.DEPLARK_API_KEY});
  if (command === 'connect') {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 388)May include surrounding context.

js
if (flags['--api-key'] && flags['--browser']) throw problem('CHOOSE_ONE_AUTH_METHOD');
    if (flags['--api-key']) return client.connectExisting(await secretInput(), flags['--verification']);
    if (!flags['--browser']) return client.connectSaved(flags['--verification']);
    if (process.platform !== 'darwin') throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY');
    return client.connectBrowser({verification: flags['--verification'], onAuthorization: async ({url}) => {
      process.stderr.write('Confirm account and project permissions: ' + url + '\n');
      if (!flags['--no-open']) { const child = spawn('/usr/bin/open', [url], {stdio: 'ignore'}); child.on('error', () => {}); child.unref(); }

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 114)May include surrounding context.

js
child.stdin.on('error', () => {}); child.stdin.end(input || '');
  });
}
export function keychainStore({origin, project, agent, accountScope=false, run = nativeCommand, keychain = null} = {}) {
  const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 114)May include surrounding context.

js
child.stdin.on('error', () => {}); child.stdin.end(input || '');
  });
}
export function keychainStore({origin, project, agent, accountScope=false, run = nativeCommand, keychain = null} = {}) {
  const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 119)May include surrounding context.

js
const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {
      try { const value = (await run('/usr/bin/security', ['find-generic-password', '-a', account, '-s', service, '-w', ...(keychain ? [keychain] : [])])).trim(); return pattern.test(value) ? value : null; }

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 119)May include surrounding context.

js
const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {
      try { const value = (await run('/usr/bin/security', ['find-generic-password', '-a', account, '-s', service, '-w', ...(keychain ? [keychain] : [])])).trim(); return pattern.test(value) ? value : null; }

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 122)May include surrounding context.

js
const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {
      try { const value = (await run('/usr/bin/security', ['find-generic-password', '-a', account, '-s', service, '-w', ...(keychain ? [keychain] : [])])).trim(); return pattern.test(value) ? value : null; }
      catch (error) { if (error.code === 'LOCAL_CLIENT_TIMEOUT') throw error; return null; }
    },
    async set(token) {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 122)May include surrounding context.

js
const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {
      try { const value = (await run('/usr/bin/security', ['find-generic-password', '-a', account, '-s', service, '-w', ...(keychain ? [keychain] : [])])).trim(); return pattern.test(value) ? value : null; }
      catch (error) { if (error.code === 'LOCAL_CLIENT_TIMEOUT') throw error; return null; }
    },
    async set(token) {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 379)May include surrounding context.

js
const account=manifest?.scope==='account'||flags['--account'];if(account&&!flags['--scope'])scope='personal';
  if (command === 'uninstall') return uninstallSkill({agent, scope, name, confirm: flags['--confirm']});
  if(account)return runAccountCLI(command,flags,{origin,agent,scope,manifest,name,runtime});
  const store = process.platform === 'darwin' ? keychainStore({origin, project, agent}) : {
    async get() { return process.env.DEPLARK_API_KEY || null; },
    async set(value) { if (value !== process.env.DEPLARK_API_KEY) throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY'); }, async delete() {},
  };

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 117)May include surrounding context.

js
export function keychainStore({origin, project, agent, accountScope=false, run = nativeCommand, keychain = null} = {}) {
  const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 118)May include surrounding context.

js
const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 118)May include surrounding context.

js
const account = sha256(endpointOrigin(origin) + '\n' + (accountScope?'account':project) + '\n' + clientName(agent)), service = accountScope?'com.deplark.account-skills':'com.deplark.skills';
  const pattern=accountScope?accountKeyPattern:keyPattern;
  // A custom keychain is only an injected test destination; never alter the search list.
  if (keychain && /[\r\n"\\]/.test(keychain)) throw problem('KEYCHAIN_PATH_INVALID');
  const suffix = keychain ? ` "${keychain}"` : '';
  return {
    async get() {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 132)May include surrounding context.

js
// Interactive security may exit zero after a failed command; verify the write.
      const saved = await this.get(); if (saved !== token) throw problem('KEYCHAIN_SAVE_FAILED');
    },
    async delete() { try { await run('/usr/bin/security', ['delete-generic-password', '-a', account, '-s', service, ...(keychain ? [keychain] : [])]); } catch {} },
  };
}
function safeConnection(value) {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 132)May include surrounding context.

js
// Interactive security may exit zero after a failed command; verify the write.
      const saved = await this.get(); if (saved !== token) throw problem('KEYCHAIN_SAVE_FAILED');
    },
    async delete() { try { await run('/usr/bin/security', ['delete-generic-password', '-a', account, '-s', service, ...(keychain ? [keychain] : [])]); } catch {} },
  };
}
function safeConnection(value) {

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 311)May include surrounding context.

js
,env:{...process.env,DEPLARK_API_KEY:'',DEPLARK_DEPLOY_TOKEN:grant.token}});child.on('error',()=>reject(problem('DEPLOY_CLIENT_UNAVAILABLE')));child.on('close',done);});if(code!==0)throw problem('DEPLOYMENT_FAILED');return {status:'deployment-client-completed',project_id:grant.project_id};}finally{await rm(temporary,{recursive:true,force:true});}
}
async function runAccountCLI(command,flags,{origin,agent,scope,manifest,name,runtime={}}){
  const store=runtime.store||(process.platform==='darwin'?keychainStore({origin,agent,accountScope:true}):{get:async()=>process.env.DEPLARK_API_KEY||null,set:async value=>{if(value!==process.env.DEPLARK_API_KEY)throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY');},delete:async()=>{}});
  const client=createAccountSkillClient({origin,agent,store,pendingStore:pendingAccountAuthorization({origin,agent,home:runtime.home}),apiKey:runtime.apiKey??process.env.DEPLARK_API_KEY,registrationSource:flags['--source']||manifest?.registration_source||'skill-pac

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/deplark.mjs (reported line 311)May include surrounding context.

js
;if(code!==0)throw problem('DEPLOYMENT_FAILED');return {status:'deployment-client-completed',project_id:grant.project_id};}finally{await rm(temporary,{recursive:true,force:true});}
}
async function runAccountCLI(command,flags,{origin,agent,scope,manifest,name,runtime={}}){
  const store=runtime.store||(process.platform==='darwin'?keychainStore({origin,agent,accountScope:true}):{get:async()=>process.env.DEPLARK_API_KEY||null,set:async value=>{if(value!==process.env.DEPLARK_API_KEY)throw problem('KEYCHAIN_UNAVAILABLE_USE_DEPLARK_API_KEY');},delete:async()=>{}});
  const client=createAccountSkillClient({origin,agent,store,pendingStore:pendingAccountAuthorization({origin,agent,home:runtime.home}),apiKey:runtime.apiKey??process.env.DEPLARK_API_KEY,registrationSource:flags['--source']||manifest?.registration_source||'skill-package'});
  if(command==='status')return client.status();if(command==='projects')return client.projects();if(command==='disconnect')return client.disconnect();
  if(comm

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.