download-results

# Download Results Skill

## Purpose

## ⚠️ Configuration Protection Policy

**CRITICAL: Never modify `~/.hotvect/` configuration files.** See `CONFIG_PROTECTION_POLICY.md` for full policy.

This skill reads `~/.hotvect/config.json` for configuration but never modifies it. Work only in user-specified directories.

Automatically download SageMaker backtest results from S3 when user needs to analyze training outputs or compare algorithm performance.

## Configuration

**CRITICAL:** You (Claude) must read `~/.hotvect/config.json` at skill invocation to construct complete `hv-ext download-results` commands.

**How this works:**
- The `hv-ext download-results` CLI tool does NOT read config.json
- YOU read config.json and construct the full command line with all arguments
- Config provides defaults that can be overridden by user-specified values

**When this skill activates:**
1. Read config: `cat ~/.hotvect/config.json`
2. Parse the JSON to extract values (you can understand JSON directly)
3. Use config values for AWS operations:
   - `aws.credential_helper` → Command to run when refreshing credentials
   - `directories.output_base_dir` → Default for `--dest-base-dir` argument

4. **Fail fast** if config doesn't exist and AWS operations are needed
5. Tell user to run `/agent hotvect-setup-agent` if config is missing

**Important:** Always construct complete `hv-ext download-results` commands with ALL required arguments explicitly specified.

## When to Invoke

Invoke this skill when:
- User mentions downloading backtest results
- User asks to check SageMaker outputs
- User wants to analyze training results from S3
- User completed a backtest and needs the outputs

## Context Requirements Before Activation

**Before activating this skill, ensure you have gathered:**
1. **Date range**: From-date and to-date for results (ask user if not mentioned)
2. **S3 base prefix**: Standard location or user-specified
3. **Destination directory**: From config `directories.output_base_di