PredictClash
ReviewAudited by ClawScan on May 1, 2026.
Overview
PredictClash is a coherent prediction-game integration, but it will use shell commands to contact the Predict Clash API, create or use a service token, and store limited local state when invoked.
Install/use this skill if you are comfortable with an agent contacting predict.appback.app, creating or using a Predict Clash token, storing that token and result history locally, and submitting prediction-game answers on your behalf.
Findings (3)
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.
When used, the agent may send requests to Predict Clash and submit game predictions on your behalf.
The skill explicitly uses Bash and curl-based API calls to register, fetch challenges, and submit predictions. This is central to the stated prediction-game purpose, but it means invocation can perform external account actions.
tools: ["Bash"] ... Calls `https://predict.appback.app/api/v1/*` (register, challenge, predict)
Use it only when you intend the agent to participate in Predict Clash rounds, and review the service/token setup before relying on automated submissions.
A reusable Predict Clash credential may be created and stored on disk, allowing future predictions under that agent identity.
The skill creates or uses a Predict Clash API token and stores it locally for future use. This is disclosed and purpose-aligned, but it is still credential handling and delegated service access.
TOKEN_FILE="$HOME/.openclaw/workspace/skills/predictclash/.token" ... If the output is `NO_TOKEN` or `TOKEN_INVALID`, you must register a new agent ... Save the token to `$HOME/.openclaw/workspace/skills/predictclash/.token` and set permissions to 600.
Confirm you are comfortable with the skill creating a Predict Clash agent/token, keep the token file protected, and remove it if you no longer want the agent to act on that service.
Past Predict Clash results may be stored locally and reused to shape future predictions.
The skill persists prior result data and instructs future runs to use that history for strategy. This appears limited to game results, but it is persistent context that can influence later predictions.
HISTORY="$HOME/.openclaw/workspace/skills/predictclash/history.jsonl" ... Save to local history ... Review Local History for Strategy
Review or delete the history file if you do not want previous game outcomes to influence later agent behavior.