Agent Skill Grading Report: opencode-cli - Score 84/100 (B)

# ✅ Agent Skill Grading Report

**Score:** 84/100 | **Grade:** B

## Quick Summary of Agent Skill Grades

### Pillar Scores for Agent Skill

| Pillar | Score | Max |
|--------|-------|-----|
| Spec Compliance | 13 | 15 |
| Progressive Disclosure | 27 | 30 |
| Ease of Use | 22 | 25 |
| Writing Style | 8 | 10 |
| Utility | 17 | 20 |
| Modifiers | -3 | ±15 |

### Issues Found: 4
- 🔴 High: 0
- 🟡 Medium: 1
- 🟢 Low: 3

## 📊 Full Grading Report for Agent Skill

# Skill Evaluation Report: opencode-cli

**Links:**
- 📁 **GitHub:** [SpillwaveSolutions/opencode_cli](https://github.com/SpillwaveSolutions/opencode_cli/blob/main/SKILL.md)
- 🛒 **Marketplace:** [View on SkillzWave](https://skillzwave.ai/skill/spillwavesolutions__opencode_cli__opencode_cli__SKILL/)
- 📤 **Submit Skill:** [Submit to SkillzWave](https://skillzwave.ai/submit/) - Skills are ranked, graded, and scanned for security
- 📊 **Tracking:** [All Graded Skills](https://github.com/SpillwaveSolutions/skill_improver/issues)

**Evaluated:** 2026-01-12
**Files Reviewed:** SKILL.md, README.md, references/provider-config.md, references/local-models.md, references/integration-patterns.md, references/cloud-providers.md, references/mcp-servers.md, references/vertex-ai-setup.md
**Grading Model:** Claude (default) (via claude)

---

## Overall Score: 84/100

| Pillar | Score | Max |
|--------|-------|-----|
| Progressive Disclosure Architecture | 27 | 30 |
| Ease of Use | 22 | 25 |
| Spec Compliance | 13 | 15 |
| Writing Style | 8 | 10 |
| Utility | 17 | 20 |
| Modifiers | -3 | ±15 |

**Grade: B**

---

## Executive Summary

This skill shows solid fundamentals with a score of 84/100. Strongest area: **Progressive Disclosure Architecture** (27/30).

---

## Detailed Scores

### Progressive Disclosure Architecture (27/30)

| Criterion | Score | Max | Assessment |
|-----------|-------|-----|------------|
| Token Economy | 9 | 10 | SKILL.md is focused and concise; minimal redundancy across files; each section earns its tokens |
| Layered Structure | 9 | 10 | Clear separation: SKILL.md as overview (195 lines), 6 reference files for deep dives; excellent topic isolation |
| Reference Depth | 5 | 5 | All references exactly one level deep from SKILL.md; no nested sub-references |
| Navigation Signals | 4 | 5 | TOC in SKILL.md; clear headers in all files; reference table maps tasks to files; some files over 200 lines lack internal TOC |

### Ease of Use (22/25)

| Criterion | Score | Max | Assessment |
|-----------|-------|-----|------------|
| Metadata Quality | 9 | 10 | Name follows kebab-case; description lists 17 specific triggers and covers all major use cases |
| Discoverability | 6 | 6 | Excellent trigger coverage: 'configure opencode', 'vertex ai', 'ollama', 'opencode mcp servers', etc. |
| Terminology Consistency | 3 | 4 | Consistent 'provider/model' format throughout; minor variation in schema URLs (opencode.ai vs opencode.dev) |
| Workflow Clarity | 4 | 5 | Quick Start has numbered steps; Verify Setup is checklist; some sections could use more explicit step numbering |

### Spec Compliance (13/15)

| Criterion | Score | Max | Assessment |
|-----------|-------|-----|------------|
| Frontmatter Validity | 5 | 5 | Valid YAML with required fields |
| Name Conventions | 4 | 4 | Correct hyphen-case format |
| Description Quality | 4 | 4 | Third-person with good trigger coverage |
| Optional Fields | 0 | 2 | No optional fields used (no penalty) |

### Writing Style (8/10)

| Criterion | Score | Max | Assessment |
|-----------|-------|-----|------------|
| Voice And Tense | 3 | 4 | Mostly imperative ('Set environment variables', 'Create service account'); occasional second-person ('your project') |
| Objectivity | 3 | 3 | No marketing language; purely instructional; factual descriptions of providers |
| Conciseness | 2 | 3 | Generally dense and focused; some sections could trim introductory sentences |

### Utility (17/20)

| Criterion | Score | Max | Assessment |
|-----------|-------|-----|------------|
| Problem Solving Power | 7 | 8 | Addresses real need for headless LLM automation; covers 75+ providers; fills gap for non-Claude workflows |
| Degrees Of Freedom | 4 | 5 | Good balance: prescriptive for setup, flexible for model/provider choice; multi-provider configs supported |
| Feedback Loops | 3 | 4 | Verify Setup checklist present; troubleshooting sections in multiple files; verification commands provided |
| Examples And Templates | 3 | 3 | Complete JSON config templates for each provider; code examples for subprocess integration |

### Modifiers Applied (-3)

**Penalties:** xml_tags_in_metadata (-5), deeply_nested_references (-2)
**Bonuses:** grep_friendly_structure (+1), exemplary_examples (+2), trigger_phrases_4plus (+1)

---

## Critical Issues (Top 4)

### Issue 1: Schema URL inconsistency

**Severity:** Low
**Location:** references/local-models.md:35
**Pillar Affected:** Ease of Use

**Problem:** Schema URL differs from SKILL.md (opencode.dev vs opencode.ai)

**Current:**
```
"$schema": "https://opencode.dev/schema/opencode.json"
```

**Suggested Rewrite:**
```
"$schema": "https://opencode.ai/config.json"
```

**Impact:** +1 point

---

### Issue 2: Missing TOC in long reference files

**Severity:** Medium
**Location:** references/provider-config.md
**Pillar Affected:** PDA

**Problem:** 380-line file lacks table of contents for navigation

**Current:**
```
No TOC, starts directly with content
```

**Suggested Rewrite:**
```
Add TOC after first heading with links to main sections
```

**Impact:** +1 point

---

### Issue 3: Minor second-person usage

**Severity:** Low
**Location:** SKILL.md:various
**Pillar Affected:** Writing Style

**Problem:** Occasional 'your' usage in otherwise imperative text

**Current:**
```
Set these before running OpenCode:
```

**Suggested Rewrite:**
```
Set these environment variables before running OpenCode:
```

**Impact:** +0.5 points

---

### Issue 4: Redundant Vertex AI section

**Severity:** Low
**Location:** SKILL.md:122-124
**Pillar Affected:** PDA

**Problem:** Vertex AI Setup section just links to reference; could merge into Reference Guides table

**Current:**
```
## Vertex AI Setup\n\nSee [vertex-ai-setup.md]...
```

**Suggested Rewrite:**
```
Remove standalone section; reference table already provides this link
```

**Impact:** +0.5 points

---

## General Recommendations

1. Add trigger phrases to description for discoverability
2. Add table of contents for files over 100 lines

---

## Grade Scale

| Grade | Score | Description |
|-------|-------|-------------|
| A | 90-100 | Production-ready |
| B | 80-89 | Good, minor work |
| C | 70-79 | Adequate, gaps |
| D | 60-69 | Needs work |
| F | <60 | Major revision |

---

---

## About This Report

This evaluation uses the [Claude Skills Best Practices](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices).

**Powered by:**
- [SkillzWave](https://skillzwave.ai) - Claude Skills Marketplace
- [SpillWave](https://spillwave.com) - AI Solutions

*Report generated for [SpillwaveSolutions/opencode_cli](https://github.com/SpillwaveSolutions/opencode_cli/blob/main/SKILL.md)*

## JSON Output

```json
{
  "skill_name": "opencode-cli",
  "evaluated_at": "2026-01-12T20:40:42.128092",
  "files_reviewed": [
    "SKILL.md",
    "README.md",
    "references/provider-config.md",
    "references/local-models.md",
    "references/integration-patterns.md",
    "references/cloud-providers.md",
    "references/mcp-servers.md",
    "references/vertex-ai-setup.md"
  ],
  "scores": {
    "spec_compliance": {
      "total": 13,
      "max": 15,
      "breakdown": {
        "frontmatter_validity": {
          "score": 5,
          "max": 5,
          "assessment": "Valid YAML with required fields"
        },
        "name_conventions": {
          "score": 4,
          "max": 4,
          "assessment": "Correct hyphen-case format"
        },
        "description_quality": {
          "score": 4,
          "max": 4,
          "assessment": "Third-person with good trigger coverage"
        },
        "optional_fields": {
          "score": 0,
          "max": 2,
          "assessment": "No optional fields used (no penalty)"
        }
      }
    },
    "pda": {
      "total": 27,
      "max": 30,
      "breakdown": {
        "token_economy": {
          "score": 9,
          "max": 10,
          "assessment": "SKILL.md is focused and concise; minimal redundancy across files; each section earns its tokens"
        },
        "layered_structure": {
          "score": 9,
          "max": 10,
          "assessment": "Clear separation: SKILL.md as overview (195 lines), 6 reference files for deep dives; excellent topic isolation"
        },
        "reference_depth": {
          "score": 5,
          "max": 5,
          "assessment": "All references exactly one level deep from SKILL.md; no nested sub-references"
        },
        "navigation_signals": {
          "score": 4,
          "max": 5,
          "assessment": "TOC in SKILL.md; clear headers in all files; reference table maps tasks to files; some files over 200 lines lack internal TOC"
        }
      }
    },
    "ease_of_use": {
      "total": 22,
      "max": 25,
      "breakdown": {
        "metadata_quality": {
          "score": 9,
          "max": 10,
          "assessment": "Name follows kebab-case; description lists 17 specific triggers and covers all major use cases"
        },
        "discoverability": {
          "score": 6,
          "max": 6,
          "assessment": "Excellent trigger coverage: 'configure opencode', 'vertex ai', 'ollama', 'opencode mcp servers', etc."
        },
        "terminology_consistency": {
          "score": 3,
          "max": 4,
          "assessment": "Consistent 'provider/model' format throughout; minor variation in schema URLs (opencode.ai vs opencode.dev)"
        },
        "workflow_clarity": {
          "score": 4,
          "max": 5,
          "assessment": "Quick Start has numbered steps; Verify Setup is checklist; some sections could use more explicit step numbering"
        }
      }
    },
    "writing_style": {
      "total": 8,
      "max": 10,
      "breakdown": {
        "voice_and_tense": {
          "score": 3,
          "max": 4,
          "assessment": "Mostly imperative ('Set environment variables', 'Create service account'); occasional second-person ('your project')"
        },
        "objectivity": {
          "score": 3,
          "max": 3,
          "assessment": "No marketing language; purely instructional; factual descriptions of providers"
        },
        "conciseness": {
          "score": 2,
          "max": 3,
          "assessment": "Generally dense and focused; some sections could trim introductory sentences"
        }
      }
    },
    "utility": {
      "total": 17,
      "max": 20,
      "breakdown": {
        "problem_solving_power": {
          "score": 7,
          "max": 8,
          "assessment": "Addresses real need for headless LLM automation; covers 75+ providers; fills gap for non-Claude workflows"
        },
        "degrees_of_freedom": {
          "score": 4,
          "max": 5,
          "assessment": "Good balance: prescriptive for setup, flexible for model/provider choice; multi-provider configs supported"
        },
        "feedback_loops": {
          "score": 3,
          "max": 4,
          "assessment": "Verify Setup checklist present; troubleshooting sections in multiple files; verification commands provided"
        },
        "examples_and_templates": {
          "score": 3,
          "max": 3,
          "assessment": "Complete JSON config templates for each provider; code examples for subprocess integration"
        }
      }
    }
  },
  "modifiers": {
    "penalties": [
      {
        "name": "xml_tags_in_metadata",
        "points": -5
      },
      {
        "name": "deeply_nested_references",
        "points": -2
      }
    ],
    "bonuses": [
      {
        "name": "grep_friendly_structure",
        "points": 1
      },
      {
        "name": "exemplary_examples",
        "points": 2
      },
      {
        "name": "trigger_phrases_4plus",
        "points": 1
      }
    ],
    "net": -3
  },
  "final_score": 84,
  "grade": "B",
  "critical_issues": [
    {
      "rank": 1,
      "title": "Schema URL inconsistency",
      "severity": "Low",
      "location": "references/local-models.md:35",
      "pillar": "Ease of Use",
      "problem": "Schema URL differs from SKILL.md (opencode.dev vs opencode.ai)",
      "current": "\"$schema\": \"https://opencode.dev/schema/opencode.json\"",
      "suggested": "\"$schema\": \"https://opencode.ai/config.json\"",
      "impact": "+1 point"
    },
    {
      "rank": 2,
      "title": "Missing TOC in long reference files",
      "severity": "Medium",
      "location": "references/provider-config.md",
      "pillar": "PDA",
      "problem": "380-line file lacks table of contents for navigation",
      "current": "No TOC, starts directly with content",
      "suggested": "Add TOC after first heading with links to main sections",
      "impact": "+1 point"
    },
    {
      "rank": 3,
      "title": "Minor second-person usage",
      "severity": "Low",
      "location": "SKILL.md:various",
      "pillar": "Writing Style",
      "problem": "Occasional 'your' usage in otherwise imperative text",
      "current": "Set these before running OpenCode:",
      "suggested": "Set these environment variables before running OpenCode:",
      "impact": "+0.5 points"
    },
    {
      "rank": 4,
      "title": "Redundant Vertex AI section",
      "severity": "Low",
      "location": "SKILL.md:122-124",
      "pillar": "PDA",
      "problem": "Vertex AI Setup section just links to reference; could merge into Reference Guides table",
      "current": "## Vertex AI Setup\\n\\nSee [vertex-ai-setup.md]...",
      "suggested": "Remove standalone section; reference table already provides this link",
      "impact": "+0.5 points"
    }
  ],
  "recommendations": [
    "Add trigger phrases to description for discoverability",
    "Add table of contents for files over 100 lines"
  ],
  "code_quality": null,
  "grading_model": "Claude (default)",
  "grading_provider": "claude"
}
```

**Links:**
- 📁 **GitHub:** [spillwavesolutions/opencode_cli](https://github.com/spillwavesolutions/opencode_cli/blob/main/SKILL.md)
- 🛒 **Marketplace:** [View on SkillzWave](https://skillzwave.ai/skill/spillwavesolutions__opencode_cli__opencode_cli__SKILL/)
- 📤 **Submit Skill:** [Submit to SkillzWave](https://skillzwave.ai/submit/) - Skills are ranked, graded, and scanned for security
- 📊 **Tracking:** [All Graded Skills](https://github.com/SpillwaveSolutions/skill_improver/issues)

---

## 📦 Recommended: Add Universal Installer Instructions

Consider adding these installation instructions to your README.md to help users install this skill across 14+ AI coding agents:

```markdown
## Installing with Skilz (Universal Installer)

The recommended way to install this skill across different AI coding agents is using the **skilz** universal installer.

### Install Skilz

```bash
pip install skilz
```

This skill supports [Agent Skill Standard](https://agentskills.io/) which means it supports 14 plus coding agents including Claude Code, OpenAI Codex, Cursor and Gemini.


### Git URL Options

```

# Install for Claude Code (your home directory)
skilz install -g https://github.com/spillwavesolutions/opencode_cli

# Or from the SkillzWave marketplace
skilz install spillwavesolutions__opencode_cli__opencode_cli
```

### Claude Code

Install to user home (available in all projects):
```bash
skilz install -g https://github.com/spillwavesolutions/opencode_cli
```

Install to current project only:
```bash
skilz install -g https://github.com/spillwavesolutions/opencode_cli --project
```

### OpenCode

Install for [OpenCode](https://opencode.ai):

```bash

# OpenCode
skilz install https://github.com/spillwavesolutions/opencode_cli --agent opencode

```

Install for Codex and Gemini too
```
# Gemini CLI
skilz install https://github.com/spillwavesolutions/opencode_cli --agent gemini


# OpenAI Codex
skilz install https://github.com/spillwavesolutions/opencode_cli --agent codex

```

Project-level install:
```bash
skilz install https://github.com/spillwavesolutions/opencode_cli --project --agent codex
```


### Install from Skillzwave Marketplace
```bash
skilz install spillwavesolutions__opencode_cli__opencode_cli --project
```

See this site [skill Listing](https://skillzwave.ai/skill/spillwavesolutions__opencode_cli__opencode_cli__SKILL/) to see how to install this exact skill to 14+ different coding agents.


### Other Supported Agents

Skilz supports 20+ coding agents including Claude Code, OpenAI Codex, OpenCode, Cursor, Gemini CLI, GitHub Copilot CLI, Windsurf, Qwen Code, Aidr, and more.

See the [skill on SkillzWave](https://skillzwave.ai/skill/spillwavesolutions__opencode_cli__opencode_cli__SKILL/) for agent-specific install commands, or check the [skilz-cli docs](https://github.com/SpillwaveSolutions/skilz-cli).


[SkillzWave](https://skillzwave.ai) is a skill marketplace for AI agents. [SpillWave](https://spillwave.com) (where I work) builds AI agent tools.
```

---

## About This Report

This evaluation uses the [Claude Skills Best Practices](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices).

**Powered by:**
- [SkillzWave](https://skillzwave.ai) - Claude Skills Marketplace
- [SpillWave](https://spillwave.com) - AI Solutions

*Report generated for [spillwavesolutions/opencode_cli](https://github.com/spillwavesolutions/opencode_cli/blob/main/SKILL.md)*

Pillar	Score	Max
Spec Compliance	13	15
Progressive Disclosure	27	30
Ease of Use	22	25
Writing Style	8	10
Utility	17	20
Modifiers	-3	±15

Pillar	Score	Max
Progressive Disclosure Architecture	27	30
Ease of Use	22	25
Spec Compliance	13	15
Writing Style	8	10
Utility	17	20
Modifiers	-3	±15

Criterion	Score	Max	Assessment
Token Economy	9	10	SKILL.md is focused and concise; minimal redundancy across files; each section earns its tokens
Layered Structure	9	10	Clear separation: SKILL.md as overview (195 lines), 6 reference files for deep dives; excellent topic isolation
Reference Depth	5	5	All references exactly one level deep from SKILL.md; no nested sub-references
Navigation Signals	4	5	TOC in SKILL.md; clear headers in all files; reference table maps tasks to files; some files over 200 lines lack internal TOC

Criterion	Score	Max	Assessment
Metadata Quality	9	10	Name follows kebab-case; description lists 17 specific triggers and covers all major use cases
Discoverability	6	6	Excellent trigger coverage: 'configure opencode', 'vertex ai', 'ollama', 'opencode mcp servers', etc.
Terminology Consistency	3	4	Consistent 'provider/model' format throughout; minor variation in schema URLs (opencode.ai vs opencode.dev)
Workflow Clarity	4	5	Quick Start has numbered steps; Verify Setup is checklist; some sections could use more explicit step numbering

Criterion	Score	Max	Assessment
Frontmatter Validity	5	5	Valid YAML with required fields
Name Conventions	4	4	Correct hyphen-case format
Description Quality	4	4	Third-person with good trigger coverage
Optional Fields	0	2	No optional fields used (no penalty)

Criterion	Score	Max	Assessment
Voice And Tense	3	4	Mostly imperative ('Set environment variables', 'Create service account'); occasional second-person ('your project')
Objectivity	3	3	No marketing language; purely instructional; factual descriptions of providers
Conciseness	2	3	Generally dense and focused; some sections could trim introductory sentences

Criterion	Score	Max	Assessment
Problem Solving Power	7	8	Addresses real need for headless LLM automation; covers 75+ providers; fills gap for non-Claude workflows
Degrees Of Freedom	4	5	Good balance: prescriptive for setup, flexible for model/provider choice; multi-provider configs supported
Feedback Loops	3	4	Verify Setup checklist present; troubleshooting sections in multiple files; verification commands provided
Examples And Templates	3	3	Complete JSON config templates for each provider; code examples for subprocess integration

Grade	Score	Description
A	90-100	Production-ready
B	80-89	Good, minor work
C	70-79	Adequate, gaps
D	60-69	Needs work
F	<60	Major revision

Agent Skill Grading Report: opencode-cli - Score 84/100 (B) #20

Description

✅ Agent Skill Grading Report

Quick Summary of Agent Skill Grades

Pillar Scores for Agent Skill

Issues Found: 4

📊 Full Grading Report for Agent Skill

Skill Evaluation Report: opencode-cli

Overall Score: 84/100

Executive Summary

Detailed Scores

Progressive Disclosure Architecture (27/30)

Ease of Use (22/25)

Spec Compliance (13/15)

Writing Style (8/10)

Utility (17/20)

Modifiers Applied (-3)

Critical Issues (Top 4)

Issue 1: Schema URL inconsistency

Issue 2: Missing TOC in long reference files

Issue 3: Minor second-person usage

Issue 4: Redundant Vertex AI section

General Recommendations

Grade Scale

About This Report

JSON Output

📦 Recommended: Add Universal Installer Instructions

Git URL Options

Claude Code

OpenCode

Install from Skillzwave Marketplace

Other Supported Agents

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions