Back to skill

Security audit

Sell The News

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward read-only wrapper for a disclosed Sell The News MCP endpoint, with the main practical risk being that query arguments are sent to that remote service.

Install this if you are comfortable sending Sell The News query parameters, tickers, searches, and related prompts to the hosted endpoint. Do not include secrets, private credentials, unpublished trading plans, or sensitive business notes in the JSON arguments.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill invokes bundled shell wrappers and therefore has code-execution capability, but the manifest does not declare any explicit tool scope such as allowed-tools or permissions. That mismatch weakens governance and review controls, making it easier for the skill to obtain broader shell/file access than readers of the manifest would expect.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The trigger conditions are broad enough to match generic requests about live financial news, summarization, direct endpoint querying, and client debugging. Overbroad routing can cause this skill to activate in more situations than necessary, increasing the chance that shell-capable wrappers are used by default and expanding the attack surface for prompt-injection or unsafe tool use.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% 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 · scripts/stn_mcp_call.sh (reported line 39)May include surrounding context.

sh
PY
)

curl -sS -D "$TMP_HEADERS" "$ENDPOINT" \
  -H 'Accept: application/json, text/event-stream' \
  -H 'Content-Type: application/json' \
  --data "$INIT_PAYLOAD" > "$TMP_BODY1"

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The script sends user-supplied tool arguments and initialization metadata to a remote MCP endpoint without any inline warning, consent prompt, or redaction step. In a skill that may be used from shell/OpenClaw for live financial queries, users may pass sensitive prompts, symbols, notes, or proprietary trading context that are silently transmitted off-host.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
96% confidence
Finding

The second curl call transmits the tool name and parsed JSON arguments to a remote endpoint, which can expose sensitive user inputs or internal context to a third party. Because this skill is explicitly designed to query a hosted Sell The News service, the context makes external transfer expected, but still risky when there is no user-facing disclosure, filtering, or restriction on what arguments may be sent.

Content

Scanner excerpt · scripts/stn_mcp_call.sh (reported line 73)May include surrounding context.

sh
CURL_ARGS+=( -H "Mcp-Session-Id: $SESSION_ID" )
fi

curl "${CURL_ARGS[@]}" --data "$CALL_PAYLOAD" > "$TMP_BODY2"

python3 - "$TMP_BODY2" <<'PY'
import json, sys

Static analysis

No suspicious patterns detected.