Builds a relationship or deal summary using the Superhuman Mail MCP server — pulling together all email history, read receipts, and calendar interactions with a specific person or company to act as a lightweight CRM. Use this skill whenever someone asks to "show me all communication with [person/company]", "what's the status of my deal with [company]", "give me a relationship summary for [person]", "when did I last talk to [person]", "pull up everything about [company]", "track this deal", "who haven't I followed up with", "show me engagement on emails I sent to [person]", "CRM view of [person]", "what's my communication history with [person]", or any variation of wanting a consolidated view of a relationship or deal. Trigger broadly — if someone wants to understand the full picture of their interactions with a person or company, this skill should activate.
80
71%
Does it follow best practices?
Impact
96%
1.41xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/deal-tracker/SKILL.mdQuality
Discovery
100%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 an excellent skill description that clearly articulates what the skill does (relationship/deal summaries from email data via Superhuman Mail MCP), when to use it (extensive list of natural trigger phrases), and how it's distinct (CRM-like consolidation of communication history). The description is thorough without being padded, uses third person voice correctly, and provides enough trigger term variety to ensure reliable activation.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: pulling email history, read receipts, calendar interactions, building relationship/deal summaries, acting as a lightweight CRM. These are concrete, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (builds relationship/deal summaries using Superhuman Mail MCP server, pulling email history, read receipts, calendar interactions) and 'when' (explicit 'Use this skill whenever...' clause with extensive trigger examples and a broad activation guideline). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'show me all communication with', 'what's the status of my deal', 'relationship summary', 'when did I last talk to', 'pull up everything about', 'track this deal', 'who haven't I followed up with', 'CRM view', 'communication history'. These are highly natural phrases. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — it's specifically about consolidated relationship/deal summaries via the Superhuman Mail MCP server. The combination of CRM-like functionality, email history aggregation, and the specific tool (Superhuman) makes it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable guidance with specific MCP tool calls and a well-structured output template, which is its primary strength. However, it is excessively verbose — the cross-platform enrichment section alone could be a separate file, and much of the descriptive text explains things Claude already knows. The lack of error handling/validation in the workflow and the monolithic structure significantly weaken the overall quality.
Suggestions
Cut the content by 50%+ — remove explanatory text like 'Internal discussions often contain context that never makes it into email' and 'Some email clients block read receipts.' Claude knows these things.
Extract the cross-platform context section (Step 2b) and the output template into separate bundle files (e.g., CROSS_PLATFORM.md and OUTPUT_TEMPLATE.md) to improve progressive disclosure and reduce the main skill's token footprint.
Add validation checkpoints: what to do if list_threads returns 0 results (try alternate search terms, broaden time window), if read receipts are unavailable, or if MCP calls fail.
Condense the output template to just the structure/headers with one brief example rather than inline commentary explaining what each section should contain.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~200+ lines. It over-explains concepts Claude already understands (how to detect MCP tools, how to interpret read receipts, what a CRM is). The cross-platform context section alone is massive and largely describes obvious search patterns. The entire Step 2b could be condensed to a few lines. Much of the output template is descriptive commentary rather than actionable structure. | 1 / 3 |
Actionability | The skill names specific MCP tool calls (Superhuman_Mail.list_threads, get_read_statuses, get_thread, create_or_update_draft, etc.) with concrete parameters and filters. The output template is detailed and copy-paste ready. Steps are concrete with specific examples of what good output looks like. | 3 / 3 |
Workflow Clarity | The multi-step workflow is clearly sequenced (Steps 1-4) with logical ordering. However, there are no validation checkpoints — no guidance on what to do if MCP calls fail, return empty results, or if the contact can't be found. For a workflow involving parallel API calls and sensitive data aggregation, error handling and verification steps are missing. | 2 / 3 |
Progressive Disclosure | The entire skill is a monolithic wall of text with no bundle files or external references. The cross-platform context section, output templates, and multi-contact view could all be split into separate reference files. Everything is inline, making the skill unnecessarily long for the context window. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
03565c9
Table of Contents
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.