Skillv1.0.4

ClawScan security

Kalshi F1 Elimination Trader · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 6, 2026, 6:12 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: This appears to be a Kalshi trading bot that mostly does what it says, but there are metadata inconsistencies and it requires a live SOLANA_PRIVATE_KEY (highly sensitive) — proceed with caution and verify before providing credentials or running live.
Guidance: Key things to check and do before installing or providing secrets: - Do not paste your primary Solana private key: create a throwaway wallet with a small amount of USDC for testing and use that for live runs if you must. - Verify the full trader.py source (the copy in the prompt was truncated). Look specifically for any network calls, hard-coded endpoints, or code that transmits data to unexpected servers. - Confirm simmer-sdk is the official package you expect (review the PyPI project and the GitHub repo linked in SKILL.md) and audit its code if you plan to run live. - Prefer to run in dry-run mode first (python trader.py) and validate the markets discovered and the logic before ever passing --live. - Because the registry metadata conflicted with the files, ask the publisher/owner to correct the package metadata so required credentials are explicit in the platform UI. - If you decide to run live, limit exposure: use a wallet with minimal funds, set conservative tunables (low max position), and monitor/log all actions. Rotate keys if you suspect anything unexpected. - If you are not comfortable auditing code or managing keys, do not provide SOLANA_PRIVATE_KEY or run with --live.

Review Dimensions

Purpose & Capability: concernThe skill's code and SKILL.md implement a Kalshi/Simmer trading bot that detects mathematically eliminated F1 drivers and (optionally) places trades. Requiring SIMMER_API_KEY and SOLANA_PRIVATE_KEY is coherent with that purpose. However, the registry summary at the top of the submission incorrectly listed 'Required env vars: none' and 'Primary credential: none', which conflicts with both SKILL.md, clawhub.json, and trader.py that require credentials. This metadata mismatch reduces trust and should be resolved before install.
Instruction Scope: noteSKILL.md and trader.py primarily describe market discovery, elimination detection using static standings, and trade execution via the simmer-sdk. The skill defaults to dry-run and only executes real trades when run with --live, which is appropriate. A notable behavior: the bundled default signal uses static championship standings (a template) — the skill suggests remixing with live F1 API data for real-time detection, so out-of-the-box results may be stale. The provided trader.py in the prompt was truncated; the missing portion could contain additional behavior, so review the full file before running.
Install Mechanism: okNo opaque download/install is present in the bundle; dependencies are standard (simmer-sdk on PyPI). clawhub.json also lists simmer-sdk as a pip requirement. There are no downloads from arbitrary URLs or archive extraction steps in the provided materials.
Credentials: concernThe skill requires SIMMER_API_KEY (API credential for Simmer) and SOLANA_PRIVATE_KEY (base58 private key used for signing on Solana). Both are logically required to place live trades, but the Solana private key is a high-value secret that grants on-chain spending authority. The skill also reads optional environment values (e.g., TRADING_VENUE, AUTOMATON_MAX_BET). Requesting a private key is proportionate for live trading, but it raises real risk — ensure you supply a limited-funds wallet and confirm you trust both the simmer-sdk package and the skill's code.
Persistence & Privilege: okThe skill is not configured as always:true and autostart is false in clawhub.json. disable-model-invocation is not set (default false), which is normal. The skill requests no system-wide config changes beyond its own config management via simmer-sdk utilities.