Content
62%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 skill with a clear workflow and useful frameworks (7-Point Filter, Tiering), but it suffers from verbosity—several sections contain general startup knowledge Claude already possesses. The actionability is limited by the lack of specificity around how Claude should actually perform investor research given its available tools. The workflow sequencing and output format specification are strong points.
Suggestions
Trim or remove the 'Common Mistakes to Avoid' and 'Angel Investor Considerations' sections—these contain general startup knowledge Claude already knows and consume significant tokens.
Clarify how Claude should actually perform investor research: specify which tools (web search, file reading) to use and what queries to run, since 'Research investors matching the criteria' is vague.
Move the 'Sourcing Investor Information' and 'Frameworks & Best Practices' detail into a referenced file (e.g., INVESTOR-RESEARCH-DETAILS.md) to keep the main skill focused on the workflow and output format.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly comprehensive but includes some content that Claude already knows (e.g., explaining what angels are, that fundraising is a funnel, basic CRM advice). The 'Common Mistakes to Avoid' and 'Angel Investor Considerations' sections contain general startup knowledge that adds token cost without much unique instructional value. However, the core frameworks (7-Point Filter, Tiering) are genuinely useful structured guidance. | 2 / 3 |
Actionability | The workflow steps are clearly described and the 7-point filter provides concrete criteria, but the skill lacks executable code or commands. The 'Build the raw list' step says 'Research investors matching the criteria' without specifying how Claude should actually do this (what tools to use, what searches to run). The output format table template is helpful but the actual research methodology is vague given Claude's tool constraints. | 2 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced with logical dependencies (read context → define criteria → build list → check conflicts → score/tier → find warm paths → deliver). The conflict-checking step serves as a validation checkpoint, and the tiering framework provides clear decision criteria. The output format is well-specified with explicit sections for conflicts and research gaps, which serves as a verification mechanism. | 3 / 3 |
Progressive Disclosure | The skill references related skills (pitch-deck, fundraising-email) and external context (startup-context.md), which is good. However, the content is quite long (~150 lines) with frameworks, best practices, common mistakes, and angel considerations all inline. The sourcing information section and common mistakes could be split into referenced files to keep the main skill leaner. | 2 / 3 |
Total | 9 / 12 Passed |