skill-research-process
Systematic process for building comprehensive Claude Code skills using parallel research agents. Triggers on "research for skill", "build skill from docs", "create comprehensive skill", or when needing to gather extensive documentation from official sources before skill creation.
86
83%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description that clearly communicates its niche purpose and provides explicit trigger conditions. Its main weakness is that the 'what' portion could be more specific about the concrete actions performed (e.g., spawning research agents, synthesizing documentation, generating SKILL.md files). The trigger terms and completeness are strong, making it easy for Claude to select appropriately.
Suggestions
Add more specific concrete actions to the 'what' portion, e.g., 'Spawns parallel research agents to gather official documentation, synthesizes findings, and generates structured SKILL.md files for Claude Code.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain ('building comprehensive Claude Code skills') and mentions 'parallel research agents' and 'gather extensive documentation from official sources', but doesn't list multiple concrete actions beyond the general process description. It's more process-oriented than action-specific. | 2 / 3 |
Completeness | Clearly answers both 'what' (systematic process for building comprehensive Claude Code skills using parallel research agents) and 'when' (explicit triggers listed with 'Triggers on...' clause and a contextual condition 'when needing to gather extensive documentation'). | 3 / 3 |
Trigger Term Quality | Includes natural trigger phrases like 'research for skill', 'build skill from docs', 'create comprehensive skill', and contextual triggers like 'gather extensive documentation from official sources'. These are terms a user would naturally say when needing this capability. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche combining skill creation, parallel research agents, and documentation gathering. The specific focus on 'Claude Code skills' and 'parallel research agents' makes it unlikely to conflict with generic coding or documentation skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, highly actionable skill that clearly defines a multi-stage research workflow with strong validation gates and error recovery. Its main weaknesses are moderate verbosity from repeated principles across multiple formats (tables, checklists, inline) and a deeply nested external reference path. The workflow clarity is excellent with explicit gates, feedback loops, and fallback strategies for common failure modes.
Suggestions
Consolidate the repeated citation/verification requirements—they appear in Quality Gate 2, Key Principles table, and Success Checklist. Keep the detailed version in one place and reference it from others.
Shorten or move the 'Agent Team Alternative for Stage 2' section to a separate reference file, as it's a conditional alternative that adds ~30 lines to the main skill body.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some sections that could be tightened—the Agent Team Alternative section adds significant length for what is essentially a conditional recommendation, and some principles are repeated across tables, checklists, and inline text (e.g., citation requirements appear in at least 3 places). | 2 / 3 |
Actionability | The skill provides concrete, executable commands (init_skill.py, package_skill.py), specific agent launch syntax, exact citation formats, clear checklist items, and a well-defined MCP tool selection table. The guidance is specific enough to follow without ambiguity. | 3 / 3 |
Workflow Clarity | The three-stage process is clearly sequenced with explicit quality gates between each stage, including validation checklists, error recovery procedures with specific fallback strategies, and feedback loops (e.g., 'If categories overlap: merge before Stage 2', 'If citation missing: add source or mark NOT_VERIFIED'). The ASCII diagram at the top provides an excellent overview. | 3 / 3 |
Progressive Disclosure | The skill references external files (agent-prompts.md, mcp-tools.md, gaps-analysis.md) appropriately, but since no bundle files were provided, we cannot verify these references resolve. The Agent Teams section references a deeply nested path (./../../../plugins/plugin-creator/skills/claude-skills-overview-2026/resources/agent-teams.md) which is a 3+ level relative path that could be fragile. The main SKILL.md itself is well-structured but somewhat long for an overview document. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- Jamie-BitFlight/claude_skills
- Commit
b9f32ec
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.