← Back to Skills Marketplace
Trade History
by
Indra Riswana
· GitHub ↗
· v1.0.0
470
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install trade-history
Description
Read and display recent trade history from local JSONL log file. Use when asked about past trades, trade recap, performance review, or to list recorded signals.
Usage Guidance
Do not run this skill until you verify the referenced Python script and the log path. Specifically: (1) Inspect the file ~/.npm-global/lib/node_modules/openclaw/skills/trade-history/read.py — do not execute it if you cannot review its contents; (2) Confirm or change the hard-coded log path (/home/windows_11/...) to a location you control, or require the skill accept a user-specified path; (3) Prefer a skill that includes its small read script in the package so you can review the code before execution; (4) If you must run it, run the script in a safe environment (non-production account, limited filesystem permissions) and back up the trades.jsonl file first. The main concerns are the missing script (unknown behavior) and the hard-coded absolute path that may read unexpected files.
Capability Analysis
Type: OpenClaw Skill
Name: trade-history
Version: 1.0.0
The skill is designed to read and display local trade history from a `trades.jsonl` file. The `SKILL.md` instructs the agent to execute a local Python script (`read.py`) with optional JSON arguments. There are no instructions for data exfiltration, unauthorized network calls, persistence, or any other malicious activity. Crucially, the 'Guardrails' section explicitly states, 'Never delete or modify the log file — read only,' indicating a clear intent for benign, read-only operation. While the `read.py` script itself (not provided) could potentially contain vulnerabilities, the instructions given to the agent in `SKILL.md` do not demonstrate malicious intent or prompt injection to achieve harmful objectives.
Capability Assessment
Purpose & Capability
Stated purpose (read/display local JSONL trade log) is plausible and the only declared runtime dependency is python3, which is proportional. However the SKILL.md hard-codes a specific log location (/home/windows_11/.openclaw/...) and a path to a read.py under ~/.npm-global/lib/node_modules/openclaw/skills/trade-history/read.py that is not part of this skill package — that is inconsistent with an instruction-only skill which should either operate on files already present or include the small script it needs.
Instruction Scope
The instructions tell the agent to execute a local Python script at a specific global path. The skill bundle contains no code to review, so the actual behavior of that script is unknown — it could read only the trades file or perform arbitrary actions. The SKILL.md also embeds a user-specific absolute log path rather than a configurable or relative path, limiting portability and raising the chance of unintended file access.
Install Mechanism
No install spec (instruction-only) is lower risk in general, but here it's problematic: the instructions assume a particular script is already installed under ~/.npm-global/lib/node_modules/... even though the registry package includes no files. That implicit dependency makes behavior environment-dependent and unverifiable.
Credentials
The skill requests no environment variables or credentials, which is appropriate. However, it references a hard-coded home-directory path and a global npm path; executing an unknown script with access to the filesystem without explicit bounds is disproportionate to the simple task of reading a JSONL file. The SKILL.md's guarantee (read-only, never modify) cannot be verified from the available materials.
Persistence & Privilege
The skill does not request 'always: true' and is user-invocable only; it does not request persistent privileges or modify other skills. Autonomous invocation remains possible (platform default) but is not combined here with elevated privileges.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install trade-history - After installation, invoke the skill by name or use
/trade-history - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: Read and display recent trade history from a local JSONL log file.
- Reads trade events from a local trades.jsonl file and displays them in reverse chronological order.
- Supports filtering trades by symbol and limiting the number of results (default 20).
- Presents each trade with details: symbol, side, entry, stop loss, take profit, note, and timestamp.
- Returns concise, structured JSON output including total trade count and items list.
- Handles missing log files gracefully by returning an empty result without errors.
- Read-only access; never modifies or deletes the log file.
Metadata
Frequently Asked Questions
What is Trade History?
Read and display recent trade history from local JSONL log file. Use when asked about past trades, trade recap, performance review, or to list recorded signals. It is an AI Agent Skill for Claude Code / OpenClaw, with 470 downloads so far.
How do I install Trade History?
Run "/install trade-history" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Trade History free?
Yes, Trade History is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Trade History support?
Trade History is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Trade History?
It is built and maintained by Indra Riswana (@newbienodes); the current version is v1.0.0.
More Skills