Daily News Short

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 7:16 AM UTC
• Version: 1.1.0

Overview

This skill is a straightforward Revid integration that sends a user-provided news topic to Revid to generate a short video, with expected but important outbound API and optional publishing considerations.

Install only if you are comfortable sending topics/prompts to Revid and using a REVID_API_KEY. Review generated videos before enabling cron or any publish-now workflow, especially for public social accounts or regulated/sensitive topics.

A.I.G

Vulnerability Patterns

• 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
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 94% confidence
• Finding: The skill includes shell-based automation examples and operational behavior that performs external API calls, but it declares no explicit tool scope or allowed-tools boundary. In an agent environment, this weakens least-privilege controls and can let the skill be invoked with broader execution capability than users expect.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill is designed to fetch live news from external sources and explicitly suggests automated publishing to connected social accounts, but it does not present a clear user-facing warning or confirmation step for network activity and autonomous posting. This creates a risk of surprising data transmission and unintended public actions if run in an automated workflow.

External Transmission

Medium

• Category: Data Exfiltration

Content

```bash
# daily-news.sh
TOPIC="${1:?topic required}"
PID=$(curl -fsS https://www.revid.ai/api/public/v3/render \
  -H "Content-Type: application/json" -H "key: $REVID_API_KEY" \
  -d "$(jq -n --arg p "$TOPIC" '{
    workflow:"article-to-video",

• Confidence: 95% confidence
• Finding: The example script transmits user-supplied topic data and an API key to an external service, which is expected for this integration but still constitutes outbound data flow. In the context of an agent skill, this is security-relevant because there is no accompanying warning about what data leaves the environment, and the example encourages unattended automation.

Natural-Language Policy Violations

Medium

• Confidence: 95% confidence
• Finding: The natural-language setting hardcodes the voice language to en-US, which can violate language or locale policy when no user preference or opt-in is provided. There is no indication in this file that the locale is justified as region-specific or that users can select an alternative.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: This shell script accesses a sensitive credential via REVID_API_KEY and uses it in outbound HTTP requests to revid.ai. Although comments show usage, the script provides no runtime warning, confirmation, or explicit user disclosure that a credential and user-supplied topic are being sent to a third-party service.

External Transmission

Medium

• Category: Data Exfiltration

Content

HERE="$(cd "$(dirname "$0")" && pwd)"
PAYLOAD=$(jq --arg p "$TOPIC" '.source.prompt=$p' "$HERE/ai-tools-news.json")

PID=$(curl -fsS https://www.revid.ai/api/public/v3/render \
  -H "Content-Type: application/json" -H "key: $REVID_API_KEY" \
  -d "$PAYLOAD" | jq -r .pid)
echo "pid=$PID"

• Confidence: 70% confidence
• Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Natural-Language Policy Violations

Low

• Confidence: 93% confidence
• Finding: The skill sets `"language": "en-US"` in the voice configuration, which forces a specific language/locale in natural-language behavior. There is no nearby note that this is optional, configurable, or limited to an English-only use case.

Vague Triggers

Low

• Confidence: 84% confidence
• Finding: This JSON manifest-style file describes a skill workflow but does not include any explicit trigger phrases, scope limits, or exclusion conditions indicating when it should or should not be invoked. For manifest files, the absence of activation specificity can make invocation behavior ambiguous if this file is used as part of skill selection or routing.

Static analysis

No suspicious patterns detected.