Asset Library Skill

• Outcome: Review
• Latest audit: Sep 16, 2026 · 4:51 AM UTC
• Version: 1.0.0

Overview

This personal document-library skill is mostly coherent, but it needs review because some code can send documents and API keys to configurable or unvalidated network destinations.

Review this before installing if you will process real personal or business documents. Use dedicated low-privilege API keys, keep CAIXU_AGENT_BASE_URL unset unless you fully trust and validate the endpoint, avoid parser export mode until download URL validation and credential stripping are fixed, and point ingestion only at directories you intend to index because extracted text and file paths are stored in a local SQLite database.

A.I.G

Vulnerability Patterns

• Insecure Skill Coding Practices: Finds exploitable flaws such as hardcoded secrets or command injection
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (2)

T09 · Insecure Skill Coding Practices

Error

• Location: caixu-shared-core/packages/skill-runner/src/index.ts:2017

Finding

Configurable model endpoint can receive API credentials and sensitive document-derived content

Content

View full analysis

0 ? { max_tokens: Math.trunc(maxTokens ?? 0) } : {}), ...(responseFormat ? { response_format: { type: responseFormat } } : {}), ...(thinkingMode ? { thinking: { type: thinkingMode } } : {}), messages: [ { role: "system", content: systemPrompt }, { role: "user", content: userPrompt } ] }) } }); ``` ```ts const apiKey = process.env.CAIXU_AGENT_API_KEY?.trim() || process.env.ZHIPU_API_KEY?.trim(); return createOpenAICompatibleSkillModelClient({ apiKey, model: process.env.CAIXU_AGENT_MODEL?.trim() || "glm-4.6", baseUrl: process.env.CAIXU_AGENT_BASE_URL?.trim(), timeoutMs: Number.isFinite(timeoutMs) && timeoutMs > 0 ? timeoutMs : 30000, httpMaxAttempts: Number.isFinite(httpMaxAttempts) && httpMaxAttempts > 0 ? httpMaxAttempts : 4, httpBaseDelayMs: Number.isFinite(httpBaseDelayMs) && ...[truncated 3589 chars]

Remediation

View remediation

T09 · Insecure Skill Coding Practices

Error

• Location: caixu-ocr-mcp/src/tools/zhipu-file-parser.ts:275

Finding

Parser-provided download URL can receive the Zhipu Bearer credential and enable SSRF

Content

View full analysis

void; }): Promise<{ assets: ParserExportAsset[]; bundleText: string | null }> { const retryConfig = getZhipuHttpRetryConfig(); let response: Response; try { response = await fetchWithZhipuRetry({ scope: `zhipu-parser:${input.apiKey.slice(-8)}`, url: input.downloadUrl, timeoutMs: requestTimeoutMs, maxAttempts: retryConfig.maxAttempts, baseDelayMs: retryConfig.baseDelayMs, maxDelayMs: retryConfig.maxDelayMs, minIntervalMs: retryConfig.minIntervalMs, onEvent: input.onEvent, label: "Zhipu parser export download", init: { headers: { Authorization: `Bearer ${input.apiKey}` } } }); } catch (error) { throw new ParsePipelineError({ code: "ZHIPU_PARSER_EXPORT_FAILED", message: error instanceof Error ? error.message : "Failed to download parser export bundle", retryable: true }); } ``` ### Techni ...[truncated 2626 chars]

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (163)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

Direct creation and mutation of libraries, parsed files, asset cards, package/lifecycle records, and maintenance operations contradict the documented non-operational role. This broad hidden authority increases the chance of accidental corruption, unauthorized data retention, or misuse of sensitive library contents under a deceptively safe label.

Content

No source excerpt is available for this finding.