Back to skill

Security audit

Geo Opt Coordinator

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent GEO API coordinator, but users should protect the local API key file it asks them to store.

Install only if you trust the GEO service and need this workflow. Store the GEO key with restrictive permissions, avoid printing it in logs or shell history, and rotate or revoke it if the file may have been exposed.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:17
Finding

API Credential Stored in a Plaintext File Without Required Access Controls

Content
View full analysis
/dev/null || cat ~/.openclaw/geo-api-key 2>/dev/null) curl -s "https://ai.gaobobo.cn/api/geo/optimization/tasks" -H "Authorization: Bearer $GEO_KEY" ``` ``` ```markdown 读取或索要密钥 → `POST /api/geo/verify-key` → 保存到 `~/.qclaw/geo-api-key`。 ``` ### Technical Analysis The Skill instructs the agent to save a reusable GEO API key directly in a plaintext file and later use it as a Bearer token. It does not require creation of the parent directory with restrictive permissions, creation of the credential file with mode `0600`, validation of existing ownership and permissions, or use of an operating-system credential store. Consequently, the credential file may inherit an unsafe process `umask` or pre-existing permissive permissions. On a shared host, or after compromise of another process running under an account that can read the file, the API key could be recovered and replayed. The issue is an insecure credential-storage practice; the audited content does not demonstrate that the file is always exposed. ### Attack Path 1. A user follows the Skill and supplies a GEO API key. 2. The agent saves the reusable key to `~/.qclaw/geo-api-key` as instructed. 3. The destination directory or file has permissive inherited or pre-existing permissions because the Skill does not enforce access controls. 4. Another local user or compromised process with read access opens the credential file. 5. The attacker extracts the key and submits it in an `Authorization: Bearer` header to `https://ai.gaobobo.cn`. 6. The attacker accesses GEO API resources allowed by the stolen key until it is revoked or expires. This path re ...[truncated 802 chars]
Remediation
View remediation
"$HOME/.qclaw/geo-api-key" ``` 3. Verify that the file is a regular file owned by the current user and is not a symbolic link before reading or writing it. 4. Reject credential files accessible by group or other users, and document the required `0600` file and `0700` directory permissions. 5. Avoid printing the key, including it in diagnostic output, or passing it through command-line arguments that may be observable in process listings. 6. Use narrowly scoped, short-lived credentials where supported, and document key rotation and immediate revocation procedures. 7. Treat HTTP error bodies and debug logs as sensitive because they may accidentally include authentication data. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill explicitly instructs reading and saving a long-lived API key from predictable local filesystem paths without warning about file permissions, local compromise risk, or safer secret-handling options. In an agent/automation context, this increases the chance of credential exposure through weak permissions, backups, logs, shell history, or other local users/processes, enabling unauthorized access to the GEO API.

Content

No source excerpt is available for this finding.