Tradebot Ops
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill is coherent for trading-bot operations, but it authorizes stopping and restarting a live trading bot without clearly requiring user approval or bounding the affected environment.
Use this only if you are comfortable with the agent diagnosing and potentially restarting your trading bot. Before installing, add or require a rule that the agent must ask for confirmation before stopping, clearing, or restarting anything connected to live trading.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A mistaken diagnosis or overly broad process/API action could interrupt or resume live trading activity unexpectedly.
This directs the agent to stop and restart a trading bot process based on detected staleness, but does not state that the user must approve the action or define exact safe command/API boundaries.
If stale: stop bot PID (runtime or heartbeat pid) → apply+restart.
Require explicit user approval before stop/restart actions, restrict operations to a named project/environment, and define safe verification and rollback steps.
A bad health check could propagate into live bot state changes, potentially affecting trades, positions, or downtime.
The skill operates on a live trading bot, where automated restart decisions can affect downstream trading behavior if stale signals or heartbeat data are misread.
Operate and monitor the trading bot safely: detect stalls, verify LIVE/halting state, validate data freshness, restart cleanly
Add containment rules such as read-only diagnosis by default, paper-trading/test-environment checks, explicit account/environment identification, and confirmation before live changes.