Simmer Skill Builder

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: simmer-skill-builder Version: 1.2.3 The bundle is a developer tool designed to assist an AI agent in generating, validating, and publishing trading skills for the Simmer platform. It contains utility scripts for account management (scripts/status.py) and a validator (scripts/validate_skill.py) that enforces best practices in generated code. The instructions in SKILL.md explicitly mandate safety features such as dry-run defaults, mandatory safeguards, and the use of the official Simmer SDK rather than direct API calls. No indicators of data exfiltration, malicious obfuscation, or unauthorized persistence were found; all behaviors are consistent with the stated purpose of a skill-building framework.

Findings (0)

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.

What this means

A malicious or low-quality tweet or web page could influence the generated bot’s trading logic if the output is not reviewed.

Why it was flagged

The builder intentionally uses user-provided social/web content and external docs to shape generated trading code. This is purpose-aligned, but untrusted text could include misleading strategy logic or prompt-like instructions.

Skill content
They might: ... Paste a tweet or thread about a strategy ... Ask your human to provide the relevant API docs, or web-fetch them if you have access.
Recommendation

Treat external strategy text and fetched docs as data, not authority. Inspect generated SKILL.md, clawhub.json, and Python code before installing or running it.

What this means

If a generated skill is run with live trading enabled, it can spend real funds or sell positions according to its strategy logic.

Why it was flagged

The generated scripts are designed to support live trading, but the template makes dry-run the default and requires an explicit --live flag for real trades.

Skill content
Usage:
    python <script>.py              # Dry run
    python <script>.py --live       # Real trades
Recommendation

Run generated skills in dry-run/paper mode first, set conservative position sizes and max trades, and only use --live after reviewing the code and strategy.

What this means

Anyone or any generated code with the API key may be able to access Simmer account data and, depending on configuration, perform trading actions.

Why it was flagged

The skill ecosystem expects a Simmer API key for account access. This is consistent with trading/account-status functionality, but it is a sensitive credential.

Skill content
"requires": { "env": [ "SIMMER_API_KEY" ], "pip": [ "simmer-sdk" ] }
Recommendation

Use a dedicated API key if possible, keep it out of generated files, avoid sharing logs, and revoke or rotate it if a generated skill behaves unexpectedly.

What this means

Future dependency changes could affect generated skills’ behavior or compatibility.

Why it was flagged

The required Python dependency is disclosed and central to the purpose, but it is not version-pinned in the artifact.

Skill content
"pip": [ "simmer-sdk" ]
Recommendation

For production trading, consider pinning tested dependency versions and reviewing dependency updates before running live.

What this means

Persistent or remotely tuned settings could change how a bot trades across runs, including in live mode.

Why it was flagged

Generated skills can use persistent local config and automaton-provided tuning to influence thresholds, sizing, and other behavior.

Skill content
Config priority: `config.json > automaton tuning > env vars > defaults`. When `slug` is provided, `load_config` automatically fetches tuned config from the Simmer Automaton
Recommendation

Review config.json, automaton tuning, and declared tunables before enabling automation or live trading.