auto-research
Deep strategic research engine — decomposes questions into parallel research threads, spawns multiple agents, and synthesizes into actionable strategic analysis
43
43%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/auto-research/SKILL.mdQuality
Discovery
32%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description communicates a high-level architectural approach (multi-agent, parallel research) but reads more like a marketing tagline than a functional skill description. It lacks a 'Use when...' clause, concrete output types, and natural user-facing trigger terms. The implementation details (spawning agents, parallel threads) don't help Claude decide when to select this skill.
Suggestions
Add an explicit 'Use when...' clause with natural trigger terms like 'deep dive', 'market research', 'competitive analysis', 'strategic question', 'comprehensive research report'.
Replace implementation details ('spawns multiple agents', 'parallel research threads') with concrete user-facing capabilities and outputs, e.g., 'Produces comprehensive research reports covering market trends, competitive landscapes, and strategic recommendations'.
Specify the types of research questions this handles to distinguish it from simpler research or Q&A skills, e.g., 'Use for complex, multi-faceted strategic questions requiring synthesis across multiple domains, not for simple factual lookups'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names some actions like 'decomposes questions into parallel research threads', 'spawns multiple agents', and 'synthesizes into actionable strategic analysis', but these are more architectural/process descriptions than concrete user-facing capabilities. It doesn't specify what kinds of research or what outputs are produced. | 2 / 3 |
Completeness | Describes what it does (decomposes questions, spawns agents, synthesizes analysis) but completely lacks a 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill. Per the rubric, a missing 'Use when...' clause caps completeness at 2, and the 'what' is also somewhat vague, so this scores a 1. | 1 / 3 |
Trigger Term Quality | Contains some relevant keywords like 'research', 'strategic analysis', and 'strategic research', but misses many natural user terms like 'deep dive', 'market research', 'competitive analysis', 'investigate', 'report', or 'findings'. Terms like 'spawns multiple agents' and 'parallel research threads' are implementation details, not user-facing trigger terms. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'deep strategic research', 'parallel research threads', and 'multiple agents' provides some distinctiveness, but 'research' and 'strategic analysis' are broad enough to overlap with simpler research or analysis skills. Without explicit trigger boundaries, it could conflict with general research or analysis skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides excellent actionability and workflow clarity with a well-structured four-phase research process, concrete examples, and proper error handling. However, it is severely over-long and monolithic — the agent prompt template, output document template, and example decomposition should be extracted into separate referenced files. The verbosity undermines token efficiency, which is especially problematic for a skill that will be loaded into context alongside conversation history.
Suggestions
Extract the agent prompt template, synthesis document template, and example decomposition into separate bundle files (e.g., AGENT-PROMPT.md, OUTPUT-TEMPLATE.md, EXAMPLES.md) and reference them from the main skill.
Compress the main SKILL.md to ~80 lines by keeping only the phase overview, key decision points, quality standards, and file references — Claude can infer standard strategic analysis structure without the full document skeleton.
Remove explanatory text that Claude already knows, such as what makes a good source, what bias awareness means, or that specificity beats generality — these are general reasoning capabilities, not skill-specific instructions.
The repeated emphasis on Thread 7 (Emerging tech) appears in the decomposition framework, the agent prompt, the synthesis template, and the example — consolidate to a single mention with a clear directive.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~250+ lines. It over-explains agent prompt templates, decomposition frameworks, and output document structures in exhaustive detail. Much of this (like explaining what a strategic analysis document should contain, or how to attribute sources) is knowledge Claude already possesses. The repeated emphasis on Thread 7 (Emerging tech) and the full document template could be significantly compressed. | 1 / 3 |
Actionability | The skill provides highly concrete, executable guidance: specific agent prompt templates, exact file paths for saving outputs, named phases with clear deliverables, agent naming conventions, specific tool usage (WebSearch/WebFetch), and a complete output document structure with frontmatter. The example decomposition is detailed and copy-paste ready. | 3 / 3 |
Workflow Clarity | The four-phase workflow is clearly sequenced with explicit checkpoints: Phase 1 includes user confirmation before spawning agents, Phase 2 has clear parallel execution instructions, Phase 3 has a defined synthesis structure, and Phase 4 has save/deliver steps. Error handling includes retry logic and gap documentation. The feedback loop of 'state decomposition to user so they can course-correct' is a good validation checkpoint. | 3 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no bundle files or external references. The full agent prompt template, the complete output document structure, the example decomposition, and the quality standards are all inline. The agent prompt template and the synthesis document template alone could each be separate referenced files, dramatically reducing the main skill's length. | 1 / 3 |
Total | 8 / 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
- huytieu/COG-second-brain
- Commit
034af4c
- 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.