← Back to Skills Marketplace
329
Downloads
0
Stars
1
Active Installs
5
Versions
Install in OpenClaw
/install simmer-calibration-report
Description
Run a calibration report on any Simmer trade journal. Win rate and EV broken down by strategy, time of day, price band, and market type. Know exactly where y...
Usage Guidance
This skill appears to be a straightforward analytics script for local Simmer trade journals, but there are a few things to check before installing or supplying secrets:
- Inconsistency: the registry metadata claims no required env vars, but the bundled files (clawhub.json and SKILL.md) require SIMMER_API_KEY. Verify whether you actually need to provide an API key for your use case. If you only want to analyze a local journal file, you should be able to avoid setting the key.
- Search the full calibration_report.py (the provided snippet is truncated) for any code paths that call get_client() or other network I/O. If the script will contact Simmer's API, decide whether you trust sending that API key and whether the calls are necessary.
- If you are unsure, run the script locally in an isolated environment (container or VM) and point CALIB_JOURNAL_PATH to a copy of your journal. Do not export SIMMER_API_KEY unless you confirm the script needs it.
- The skill auto-detects a workspace path (default ~/.openclaw/workspace). If you install it, set CALIB_JOURNAL_PATH explicitly to the exact file you want analyzed to avoid unexpected file access.
- Review simmer-sdk (the dependency) or run in a virtualenv to limit side effects; inspect simmer_sdk.skill.load_config/update_config behavior so you understand whether it writes config files.
If you want, I can scan the rest of calibration_report.py (the truncated portion) for any network calls, secrets exfiltration, or calls to get_client()/SimmerClient to improve confidence.
Capability Analysis
Type: OpenClaw Skill
Name: simmer-calibration-report
Version: 1.0.4
The skill bundle is a legitimate analytical tool designed to process trade journals (JSONL files) and generate performance reports. The Python code in calibration_report.py performs standard data processing, filtering, and statistical calculations without any evidence of malicious execution, data exfiltration, or prompt injection. It uses a dedicated SDK (simmer-sdk) for configuration management and follows the expected patterns for an OpenClaw AgentSkill.
Capability Assessment
Purpose & Capability
The code and documentation focus on reading a local JSONL trade journal and producing analytics — that matches the skill description. However the packaged metadata (clawhub.json) and runtime docs ask for SIMMER_API_KEY and pip-install simmer-sdk. An API key is plausible if the skill sometimes talks to Simmer's API, but the visible report logic operates purely on local journal files and does not require network access for analysis. The declared registry metadata earlier (no required env vars) contradicts the bundled clawhub.json/SKILL.md which do list SIMMER_API_KEY.
Instruction Scope
Runtime instructions are focused and explicit: install simmer-sdk, set SIMMER_API_KEY, and point to a journal file (or let it auto-detect in a workspace path). The script reads files under a workspace path (default ~/.openclaw/workspace/data/...) and may read/update a skill config via simmer_sdk.skill.load_config/update_config. The actions described are within the stated purpose, but the instructions also reference environment variables (SIMMER_API_KEY, OPENCLAW_WORKSPACE, TRADING_VENUE) beyond the calibration inputs — this is expected for integration but worth noting.
Install Mechanism
No network download/install spec in the registry; the README/SKILL.md ask users to pip install simmer-sdk (a normal package manager operation). All bundled code is included in the skill, and there are no external download URLs or extracted archives in the manifest — low install risk from the skill itself.
Credentials
There is a proportionality mismatch: the top-level registry metadata reported no required environment variables, yet clawhub.json and the SKILL.md/README request SIMMER_API_KEY (and the code checks for it in get_client()). For purely local journal analysis, a service API key may be unnecessary. Requiring an API key (and instructing users to export it) increases risk if the key is not actually needed by the parts of the script that run automatically. Other env vars used (OPENCLAW_WORKSPACE, TRADING_VENUE, CALIB_* tuning vars) are reasonable for locating journals and configuring behavior.
Persistence & Privilege
always is false and user-invocable is true. The clawhub.json sets a cron entry and automaton.managed true (it can be scheduled), which means the skill can be run automatically by the platform if enabled — this is expected for an analytics cron. The skill does not request system-wide configuration changes or other skills' credentials from the visible code.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install simmer-calibration-report - After installation, invoke the skill by name or use
/simmer-calibration-report - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.4
AgentSkills spec frontmatter + UTF-8 stdout
v1.0.3
Republish for Simmer registry sync
v1.0.2
- Internal update to calibration_report.py; no user-facing changes.
- Documentation and usage instructions remain unchanged.
v1.0.1
simmer-calibration-report 1.0.1
- Added a new README.md for additional documentation.
- Revised SKILL.md for clearer language, improved setup instructions, and updated example output.
- Reorganized troubleshooting section for easier reference.
- Adjusted configuration and quick command sections for brevity and clarity.
v1.0.0
Initial release: Generate detailed calibration reports from any Simmer trade journal.
- Breaks down win rate and expected value by strategy, time of day, price band, and market type.
- Works with both live and simulated Simmer journals.
- Flexible configuration via environment variables or CLI.
- Supports filtering by minimum trades, lookback period, and inclusion of unresolved trades.
- Simple one-command usage with clear example output.
Metadata
Frequently Asked Questions
What is Simmer Calibration Report?
Run a calibration report on any Simmer trade journal. Win rate and EV broken down by strategy, time of day, price band, and market type. Know exactly where y... It is an AI Agent Skill for Claude Code / OpenClaw, with 329 downloads so far.
How do I install Simmer Calibration Report?
Run "/install simmer-calibration-report" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Simmer Calibration Report free?
Yes, Simmer Calibration Report is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Simmer Calibration Report support?
Simmer Calibration Report is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Simmer Calibration Report?
It is built and maintained by Dyll (@djdyll); the current version is v1.0.4.
More Skills