Back to skill

Security audit

OSINT Daily Brief

Security checks for vulnerabilities and agentic risk

Overview

This OSINT briefing skill uses disclosed public-intelligence sources and APIs, but users should treat target queries and scheduled monitoring as sensitive.

Install only if you are comfortable sending OSINT targets and search terms to third-party providers such as Tavily and Shodan, and only run it against assets or investigations you are authorized to research. Be careful with the optional cron examples because they create ongoing monitoring jobs and may forward reports outside your environment.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (9)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill is framed for 'any target' including companies, IPs, persons, and keywords, with examples such as competitor monitoring and threat-actor tracking, but without meaningful scope restrictions or exclusions. That makes it easy to use for invasive profiling, competitive surveillance, or collection on individuals without clear authorization boundaries.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill sends target identifiers and investigator queries to third-party services, but the documentation does not clearly disclose that this data leaves the local environment. Users may unknowingly expose sensitive investigation targets, internal watchlists, or customer names to Tavily and Shodan, creating privacy, confidentiality, and vendor-risk concerns.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This code posts search queries and the API key to Tavily, causing target-related intelligence requests to be transmitted off-platform. In an OSINT context this may be expected, but it is still security-relevant because investigation targets and search terms can be sensitive and may reveal monitoring activity.

Content

Scanner excerpt · SKILL.md (reported line 38)May include surrounding context.

md
def tavily_search(query: str, max_results: int = 5) -> list[dict]:
    """AI-optimized web search — returns full page content, not just snippets."""
    r = requests.post(
        "https://api.tavily.com/search",
        json={
            "api_key":     os.environ["TAVILY_API_KEY"],

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This code posts search queries and the API key to Tavily, causing target-related intelligence requests to be transmitted off-platform. In an OSINT context this may be expected, but it is still security-relevant because investigation targets and search terms can be sensitive and may reveal monitoring activity.

Content

Scanner excerpt · SKILL.md (reported line 38)May include surrounding context.

md
def tavily_search(query: str, max_results: int = 5) -> list[dict]:
    """AI-optimized web search — returns full page content, not just snippets."""
    r = requests.post(
        "https://api.tavily.com/search",
        json={
            "api_key":     os.environ["TAVILY_API_KEY"],

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

md
def tavily_search(query: str, max_results: int = 5) -> list[dict]:
    """AI-optimized web search — returns full page content, not just snippets."""
    r = requests.post(
        "https://api.tavily.com/search",
        json={
            "api_key":     os.environ["TAVILY_API_KEY"],
            "query":       query,

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The Shodan lookup sends a target IP or domain plus an API key to an external reconnaissance service, which can expose what assets are being investigated. For sensitive investigations, this may leak customer targets, incident-response focus areas, or internal interest in particular infrastructure.

Content

Scanner excerpt · SKILL.md (reported line 102)May include surrounding context.

md
return {"error": "SHODAN_API_KEY not set"}
    try:
        r = requests.get(
            f"https://api.shodan.io/shodan/host/{ip_or_domain}",
            params={"key": key},
            timeout=10
        )

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 201)May include surrounding context.

md
- Tavily free tier: 1,000 searches/month. Each run uses ~3–5 searches.
- Shodan free tier: limited to 1 result per query on some endpoints
- DNS recon requires `dig` installed: `sudo apt install dnsutils`
- WHOIS requires `whois` installed: `sudo apt install whois`
- For dark web mentions, pair with Tor proxy (Ahmia search via SOCKS5)

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 202)May include surrounding context.

md
- Tavily free tier: 1,000 searches/month. Each run uses ~3–5 searches.
- Shodan free tier: limited to 1 result per query on some endpoints
- DNS recon requires `dig` installed: `sudo apt install dnsutils`
- WHOIS requires `whois` installed: `sudo apt install whois`
- For dark web mentions, pair with Tor proxy (Ahmia search via SOCKS5)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The documentation expands the skill from ordinary web/DNS/WHOIS/Shodan OSINT into dark-web/Tor-assisted collection, which materially broadens the surveillance and legal-risk profile beyond the declared scope. Because this is presented as a simple note rather than a gated, high-risk capability with warnings and restrictions, it can enable misuse and increase operational and compliance risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.