ClawHub
Back to skill

Security audit

AgentEarth

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AgentEarth API integration that sends task queries and validated tool parameters to AgentEarth, with no local persistence or hidden installation behavior found.

Install only if you intend to use AgentEarth as an external tool marketplace. Treat queries, parameters, and tool results as data shared with AgentEarth, keep AGENT_EARTH_API_KEY in host-managed secret storage, and avoid sending secrets or sensitive personal data unless the selected AgentEarth tool requires it and you trust that use.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Memory Manipulation

High
Category
Memory Poisoning
Content
- Treat the Execute body as dynamic; only `params` is fixed.
- Always determine param keys from Recommend response `tools[*].input_schema.properties`.
- Always fill values from user input or clear conversation context.
- Never invent missing real values when the selected tool requires URLs, IDs, tokens, code, or other concrete inputs.
- Always remove unknown keys when `additionalProperties` is `false`.
- Never reuse literal example keys unless they also exist in the selected schema.
Confidence
80% confidence
Finding
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

External Transmission

Medium
Category
Data Exfiltration
Content
**Bash (Linux / macOS / WSL):**

```bash
curl -sS -X POST "https://agentearth.ai/agent-api/v1/tool/recommend" \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $AGENT_EARTH_API_KEY" \
  -d '{"query":"<task-focused natural language query>","limit":5}'
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
**Bash (Linux / macOS / WSL):**

```bash
curl -sS -X POST "https://agentearth.ai/agent-api/v1/tool/recommend" \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $AGENT_EARTH_API_KEY" \
  -d '{"query":"Math calculation tools","limit":5}'
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
**Bash (Linux / macOS / WSL):**

```bash
curl -sS -X POST "https://agentearth.ai/agent-api/v1/tool/execute?recommend_id=<id>&service_id=<id>&tool_name=<name>" \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: $AGENT_EARTH_API_KEY" \
  -d '{"params":{"<schema_field_1>":"<value>","<schema_field_2>":"<value>"}}'
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:79