Back to skill

Security audit

Kalshi Weather Trader

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-built for Kalshi weather trading, but it needs review because it asks for a raw Solana private key and can run live automated trades with bypassable safeguards.

Install only if you intentionally want automated Kalshi weather trading. Use a dedicated low-balance Solana wallet, never reuse a primary wallet key, keep dry-run mode until you understand the strategy, avoid --no-safeguards for live trades, and enable cron or quiet live mode only with explicit spending limits you can afford to lose.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill explicitly requires access to environment variables and external network services, including a Solana private key and API endpoints, yet no declared permissions are present. This creates a transparency and policy-enforcement gap: users or hosting systems may invoke a skill with sensitive capabilities without clear consent boundaries or capability review.

Vague Triggers

Medium
Confidence
75% confidence
Finding
The description says to use the skill not only for trading but also to 'check NOAA forecasts,' which broadens activation beyond clearly financial/trading intent. That can cause the agent to invoke a wallet-capable, trade-oriented skill for generic weather queries, increasing the chance of unnecessary secret exposure, confusing tool routing, or accidental progression into trading workflows.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The manifest requires a raw Solana private key to be supplied to the skill for signing trades, but the description does not provide a prominent warning about the sensitivity of that credential, custody implications, or the risk of loss if the skill or its dependencies are compromised. In a trading automation context, that omission is dangerous because users may provide a wallet key without understanding they are granting the skill direct authority to move funds and sign transactions.

External Transmission

Medium
Category
Data Exfiltration
Content
4. **Verify KYC**
   - Required for Kalshi buys (not sells)
   - Complete at [dflow.net/proof](https://dflow.net/proof)
   - Check status: `curl "https://api.simmer.markets/api/proof/status?wallet=YOUR_SOLANA_ADDRESS"`

5. **Fund the wallet**
   - SOL on Solana mainnet for transaction fees (~0.01 SOL)
Confidence
82% confidence
Finding
The skill instructs users to send wallet-linked data to external services such as api.simmer.markets and dflow/proof for KYC status and trade execution. In a financial skill handling private keys, balances, positions, and wallet identifiers, external transmission materially increases privacy, tracking, and account-targeting risk if data flows are not tightly scoped and disclosed.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
- **Market status**: Skips if market already resolved
- **Kalshi maintenance**: Kalshi's clearinghouse has a weekly maintenance window on Thursdays 3:00-5:00 AM ET — orders during this window will fail

Disable with `--no-safeguards` (not recommended).

## Troubleshooting
Confidence
88% confidence
Finding
The skill exposes a '--no-safeguards' option that disables protections against flip-flopping, slippage, time-decay, and related unsafe trading conditions. In a system capable of live financial execution, allowing an agent or user to bypass guardrails can directly increase the likelihood of harmful or unintended trades, especially if commands are composed automatically.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.