evaluate-report

# /evaluate:report View evaluation results and benchmark reports for a skill or plugin. ## When to Use This Skill | Use this skill when... | Use alternative when... | |------------------------|------------------------| | Want to see results from a past evaluation | Need to run new evaluations -> `/evaluate:skill` | | Comparing benchmark runs over time | Want improvement suggestions -> `/evaluate:improve` | | Reviewing quality trends for a plugin | Need structural compliance -> `plugin-compliance-check.sh` | ## Parameters Parse these from `$ARGUMENTS`: | Parameter | Default | Description | |-----------|---------|-------------| | `<target>` | required | `plugin/skill-name` for skill or `plugin-name` for plugin | | `--latest` | true | Show most recent benchmark only | | `--history` | false | Show all benchmark iterations | | `--compare` | false | Compare current vs previous benchmark | ## Execution ### Step 1: Resolve target Determine if `$ARGUMENTS` refers to a skill or plugin: - Contains `/`: treat as `plugin-name/skill-name`, look for `<plugin>/skills/<skill>/eval-results/benchmark.json` - No `/`: treat as `plugin-name`, look for `<plugin>/eval-results/plugin-benchmark.json` If the target file does not exist, report that no evaluation results are available and suggest running `/evaluate:skill` or `/evaluate:plugin-batch`. ### Step 2: Load results **Skill-level report** (`--latest` or default): Read `benchmark.json` and display: ``` ## Evaluation Report: <plugin/skill-name> **Last run**: <timestamp> **Runs per eval**: N | Metric | With Skill | Baseline | Delta | |--------|-----------|----------|-------| | Pass Rate | 85% | 42% | +43% | | Duration | 14s | 12s | +2s | ### Per-Eval Results | Eval ID | Description | Pass Rate | Failures | |---------|-------------|-----------|----------| | eval-001 | Basic usage | 100% | — | | eval-002 | Edge case | 67% | assertion X | ``` **Skill-level history** (`--history`): Read `history.json` and display improvement iterations: ``` ## Evaluation History: <plugin/skill-name> | Version | Date | Pass Rate | Changes | |---------|------|-----------|---------| | v3 (current) | 2026-03-04 | 85% | Improved step 3 instructions | | v2 | 2026-03-01 | 72% | Added error handling | | v1 | 2026-02-28 | 55% | Initial evaluation | ``` **Skill-level comparison** (`--compare`): Read current and previous benchmarks, show delta: ``` ## Comparison: <plugin/skill-name> | Metric | Previous | Current | Delta | |--------|----------|---------|-------| | Pass Rate | 72% | 85% | +13% | | Duration | 16s | 14s | -2s | Improved evals: eval-002, eval-004 Regressed evals: none ``` **Plugin-level report**: Read `plugin-benchmark.json` and display: ``` ## Plugin Report: <plugin-name> | Skill | Evals | Pass Rate | Status | |-------|-------|-----------|--------| | skill-a | 4 | 100% | PASS | | skill-b | 3 | 67% | NEEDS WORK | **Overall**: 82% pass rate **Skills evaluated**: N / M total ``` ## Agentic Optimizations | Context | Command | |---------|---------| | Find skill benchmark | `cat <plugin>/skills/<skill>/eval-results/benchmark.json \| jq .summary` | | Find plugin benchmark | `cat <plugin>/eval-results/plugin-benchmark.json \| jq .summary` | | Check for history | `cat <plugin>/skills/<skill>/eval-results/history.json \| jq '.iterations \| length'` | | List available results | `find <plugin> -name benchmark.json -o -name plugin-benchmark.json` | ## Quick Reference | Flag | Description | |------|-------------| | `--latest` | Most recent benchmark (default) | | `--history` | Show all iterations | | `--compare` | Compare current vs previous |