ClawHub
Back to skill

Security audit

OATDA Vision Analysis

Security checks for vulnerabilities and agentic risk

Overview

The skill has a clear image-analysis purpose, but it needs review because its executable shell template can mix user-supplied text with an API-key-bearing command unsafely.

Review before installing. Use this only when you are comfortable sending the image and prompt to OATDA, avoid confidential screenshots or documents unless approved, and prefer a safer implementation that builds the JSON payload with jq or a typed client instead of textual substitution into the provided shell command.

A.I.G

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

Error
Location
SKILL.md:57
Finding

Potential Shell Command Injection Through Unsafe User-Controlled JSON Interpolation

Content
View full analysis
/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}" && \ curl -s -X POST "https://oatda.com/api/v1/llm/image" \ -H "Content-Type: application/json" \ -H "Authorization: Bearer $OATDA_API_KEY" \ -d '{ "provider": "", "model": "", "contents": [ {"type": "text", "text": ""}, {"type": "image", "image": {"url": "", "detail": "auto"}} ] }' ``` The same unsafe construction pattern is repeated in the example at `SKILL.md`, lines 117–129. ### Technical Analysis The documented command places the request body inside a single-quoted shell argument while instructing the Agent to substitute values such as `` and ``. These values may originate from user input, but the Skill does not require them to be encoded with a JSON serializer or passed through a shell-safe mechanism. If an implementation performs direct textual substitution, a single quote in an attacker-controlled value can terminate the shell argument. The remaining input can then be interpreted as shell syntax. Even where shell execution is not achieved, unescaped double quotes, backslashes, control characters, or newlines can alter the JSON structure or produce an invalid request. This is particularly sensitive because the same shell process exports `OATDA_API_KEY`. Successful injected commands would execute in an environment containing that credential and with the privileges of the Agent process. Access to `OATDA_API_KEY` and the fallback credential file `~/.oatda/credentials.json` is functionally justified for authentication to the declared OATDA API. No unrelated credential destination was ...[truncated 1743 chars]
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

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 · SKILL.md (reported line 10)May include surrounding context.

md
"openclaw":
      {
        "emoji": "👁️",
        "requires": { "bins": ["curl", "jq"], "env": ["OATDA_API_KEY"], "config": ["~/.oatda/credentials.json"] },
        "primaryEnv": "OATDA_API_KEY",
      },
  }

Credential Access

High
Category
Privilege Escalation
Confidence
88% confidence
Finding

The skill explicitly reads an API key from a local credentials file and exports it into the process environment for later use. While intended for legitimate authentication, this is still credential access behavior and increases exposure because local secrets are being programmatically retrieved and placed into a broader execution context.

Content

Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.

All commands need the OATDA API key. Resolve it inline for each exec call:

bash
export OATDA_API_KEY="${OATDA_API_KEY:-$(cat ~/.oatda/credentials.json 2>/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}"

If the key is empty or null, tell the user to get one at https://oatda.com and configure it.

Credential Access

High
Category
Privilege Escalation
Confidence
90% confidence
Finding

This command both retrieves a local API key from disk and uses it in an authenticated outbound request. Although the destination appears to be the intended service, combining credential access with network transmission raises the risk of secret mishandling, leakage through logs, or misuse if the command is modified or invoked in an unsafe context.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

CRITICAL: The endpoint is /api/v1/llm/image (NOT /api/v1/llm/generate-image — that's for image generation). The body uses a contents array, NOT a simple prompt string.

bash
export OATDA_API_KEY="${OATDA_API_KEY:-$(cat ~/.oatda/credentials.json 2>/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}" && \
curl -s -X POST "https://oatda.com/api/v1/llm/image" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OATDA_API_KEY" \

Credential Access

High
Category
Privilege Escalation
Confidence
90% confidence
Finding

The example again shows reading a local credential file and immediately attaching the retrieved bearer token to an external request. Even as sample code, this normalizes direct credential handling patterns that can be unsafe in agent environments, where outputs, histories, or wrappers may inadvertently expose secrets.

Content

Scanner excerpt · SKILL.md (reported line 119)May include surrounding context.

User: "Describe this image: https://example.com/photo.jpg"

bash
export OATDA_API_KEY="${OATDA_API_KEY:-$(cat ~/.oatda/credentials.json 2>/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}" && \
curl -s -X POST "https://oatda.com/api/v1/llm/image" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OATDA_API_KEY" \

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest description says the skill triggers when the user wants to "analyze, describe, or understand images" and similar broad intents. These are common, high-level requests without explicit trigger boundaries or exclusion conditions, which can cause unintended invocation in ordinary conversation involving images.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The skill instructs the agent to transmit user-supplied image content and prompts to an external third-party API. Even though this is the expected function of the skill, it creates a real data egress path that could expose sensitive images, embedded text, or user data to the remote service if invoked on confidential content.

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

bash
export OATDA_API_KEY="${OATDA_API_KEY:-$(cat ~/.oatda/credentials.json 2>/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}" && \
curl -s -X POST "https://oatda.com/api/v1/llm/image" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OATDA_API_KEY" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The example command demonstrates sending image URLs and prompts to an external API endpoint, reinforcing an operational exfiltration path. In a skill context, such examples are effectively executable guidance and can lead to unintentional disclosure of private or regulated image content.

Content

Scanner excerpt · SKILL.md (reported line 120)May include surrounding context.

bash
export OATDA_API_KEY="${OATDA_API_KEY:-$(cat ~/.oatda/credentials.json 2>/dev/null | jq -r '.profiles[.defaultProfile].apiKey' 2>/dev/null)}" && \
curl -s -X POST "https://oatda.com/api/v1/llm/image" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OATDA_API_KEY" \
  -d '{

Static analysis

No suspicious patterns detected.