AI Credit Share 中国平台助手

• Outcome: Review
• Latest audit: May 28, 2026 · 9:02 AM UTC
• Version: 1.0.1

Overview

This appears to be a real AI Credit Share platform helper, but it handles live account credentials and money-affecting actions with weak safeguards.

Install only if you intend to let this skill operate a live AI Credit Share account. Use an explicit strong password, avoid the default-password path, treat ~/.aicreditshare as sensitive, restrict its file permissions, and manually confirm every publishing, hiring, acceptance, cancellation, payout, dispute, profile/webhook change, message send, or API key rotation before execution.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Rogue Agent: Self-Modification, Session Persistence
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger

Findings (15)

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: The script exposes a high-risk credential-management action, `agent regenerate-secret`, that rotates the API secret directly from a general-purpose automation client. This is dangerous because secret rotation is an irreversible administrative action that can break existing integrations and, if the skill is invoked unexpectedly or by an untrusted workflow, can cause account lockout or denial of service for the user’s automation.

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The script retrieves a default password from an environment variable or local file and falls back to a hardcoded default password of "Test123456". This creates predictable credentials and normalizes persistent password storage for an account-management script, which materially increases the chance of account compromise if the file, environment, or default value is exposed.

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The registration flow prints the password, API key, and API secret directly to stdout. Secrets shown in terminal output can be captured by shell history tools, terminal logging, screen sharing, CI logs, or shoulder surfing, exposing full account credentials.

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The script stores API keys, API secrets, tokens, email addresses, and registration state in plaintext under ~/.aicreditshare/config.json. Plaintext local credential storage significantly increases the blast radius of local compromise, backup leakage, multi-user host exposure, or accidental file disclosure.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The README advertises actions such as publishing tasks, paying rewards, hiring skills, and freezing deposits, but it does not clearly warn users that these operations can commit account funds or create financial obligations. In an agent-driven workflow, this omission increases the risk that a user triggers a costly action through automation or natural-language prompts without informed consent.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The example prompts encourage users to ask the assistant to publish tasks and claim work directly, but they do not mention that these actions may lock funds, spend balance, or create obligations on the platform. Because the skill is designed for conversational execution, omission of warnings here materially raises the chance of unintended financial actions from ambiguous or casual user requests.

Vague Triggers

Medium

• Confidence: 85% confidence
• Finding: The invocation phrases are very broad and map to common user requests like checking messages, balance, or publishing tasks without requiring explicit confirmation of account-affecting actions. In a skill that can spend funds, freeze deposits, send messages, or change account state, broad triggers increase the chance of unintended or overbroad execution.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill states it will save returned `agentApiKey` and `agentApiSecret` but provides no warning, consent flow, storage details, or handling constraints for these highly sensitive credentials. Persisting API secrets without transparent security controls increases the risk of credential theft, account takeover, and unauthorized financial or administrative actions on the platform.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill advertises API key reset as a normal feature without prominently warning that regeneration invalidates the existing secret and can break other automations or lock users out until dependent systems are updated. Because this action is destructive and affects authentication, lack of clear warning can cause accidental denial of service and account disruption.

Vague Triggers

Medium

• Confidence: 80% confidence
• Finding: The description advertises broad autonomous actions such as registration, posting tasks, accepting work, publishing skills, hiring, and checking balances without stating clear user-trigger boundaries, consent gates, or scope limitations. In an agent ecosystem, this can cause overbroad activation or user misunderstanding, increasing the chance of unintended account actions, financial operations, or platform abuse.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The `regenerate-secret` operation performs an irreversible sensitive account action with no confirmation, dry-run, or warning beyond a status message. In an agent skill context, that increases the chance of accidental or automated execution, which can immediately invalidate stored credentials and disrupt all dependent clients or jobs.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The script saves the default password to ~/.aicreditshare/default_password without meaningful warning, consent, or protection. Even if intended for convenience, silently persisting a reusable password is unsafe because users may not realize their credentials remain accessible on disk.

Missing User Warnings

Medium

• Confidence: 99% confidence
• Finding: The script explicitly echoes the user's password in cleartext during registration and login. This is highly sensitive data exposure and is unnecessary for the platform-assistant purpose; anyone with access to terminal output or logs can recover the password immediately.

Session Persistence

Medium

• Category: Rogue Agent

Content

echo -e "${YELLOW}⚠️  请立即保存以上凭证，仅显示一次！${NC}"
        
        # 保存到配置文件
        mkdir -p ~/.aicreditshare
        cat > ~/.aicreditshare/config.json << EOF
{
  "agent_id": "$agent_id",

• Confidence: 98% confidence
• Finding: The script persists agent credentials, including API secret, in a reusable config file under the user's home directory, creating durable session and credential exposure. For a platform assistant that can register agents and manage tasks, compromised local files could enable unauthorized account use and API abuse.

Session Persistence

Medium

• Category: Rogue Agent

Content

echo "================================"
        
        # 保存配置
        mkdir -p ~/.aicreditshare
        cat > ~/.aicreditshare/config.json << EOF
{
  "agent_id": "$agent_id",

• Confidence: 94% confidence
• Finding: The login flow writes the authentication token and API key to ~/.aicreditshare/config.json, extending the lifetime of authenticated sessions beyond immediate use. Persisted session material raises the risk of unauthorized reuse by local attackers, malware, backups, or other processes with filesystem access.

Static analysis

No suspicious patterns detected.