product-diagnosis
Diagnoses product health by cross-referencing Amplitude analytics (dashboards, charts, funnels, feedback, AI agent analytics), optionally Datadog (errors, latency, stack traces), and optionally Slack (qualitative feedback, bug reports, feature requests). Identifies what's broken, what's working, and what to do about it — with root causes, not just symptoms. Use when asked to "diagnose my product", "what's going on", "product health check", "what's broken", "where are users struggling", "give me a product diagnosis", or "what should I focus on".
90
88%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
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 hits all the marks. It provides specific capabilities with named tools and data types, includes a comprehensive set of natural trigger phrases in an explicit 'Use when...' clause, and occupies a clearly distinct niche combining product analytics, error monitoring, and qualitative feedback for holistic product diagnosis.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and data sources: cross-referencing Amplitude analytics (with sub-types like dashboards, charts, funnels, feedback, AI agent analytics), Datadog (errors, latency, stack traces), and Slack (qualitative feedback, bug reports, feature requests). Also specifies the output: root causes, what's broken, what's working, and what to do about it. | 3 / 3 |
Completeness | Clearly answers both 'what does this do' (diagnoses product health by cross-referencing multiple data sources, identifies root causes) AND 'when should Claude use it' with an explicit 'Use when...' clause containing multiple trigger phrases. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger phrases users would actually say: 'diagnose my product', 'what's going on', 'product health check', 'what's broken', 'where are users struggling', 'give me a product diagnosis', 'what should I focus on'. These are highly natural and varied. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — the combination of Amplitude analytics, Datadog, and Slack for product health diagnosis is a very clear niche. The specific tool names and the 'product diagnosis' framing make it unlikely to conflict with generic analytics or monitoring skills. | 3 / 3 |
Total | 12 / 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 strong, highly actionable diagnostic skill with excellent workflow clarity and specific tool call guidance across multiple data sources. Its main weakness is length — the document tries to be both a workflow guide and a reference manual in a single file, and includes some conceptual explanation (error rate vs failure rate, RICE definitions) that Claude doesn't need. Trimming explanatory content and splitting reference material into bundle files would improve token efficiency.
Suggestions
Extract the RICE scoring tables, impact/confidence/effort anchors into a separate RICE_REFERENCE.md file and link to it — this is reference material, not workflow guidance.
Remove or significantly compress the 'Core Principle' section; Claude understands that error rates don't equal failure rates and that qualitative signals matter. A single sentence reminder is sufficient.
Move the AI Agent Analytics field table and the opportunity structure template into separate bundle files to reduce the main skill's token footprint.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is thorough and mostly earns its length given the complexity of a multi-phase diagnostic workflow, but there's some unnecessary explanation (e.g., the 'Core Principle' section explaining error rate vs failure rate concepts Claude already understands, and the detailed RICE scoring anchors that are standard PM knowledge). The tables and structured phases are efficient, but the overall document could be tightened by ~20-30%. | 2 / 3 |
Actionability | Highly actionable throughout — every phase specifies exact tool calls with parameters (e.g., `search` with `isOfficial: true, sortOrder: "viewCount"`, `query_ai_sessions` with `responseFormat: "detailed"`), concrete thresholds (>15% deviation), specific search queries for Slack, and a complete output template with RICE scoring formulas. Claude knows exactly what to call and in what order. | 3 / 3 |
Workflow Clarity | Excellent 6-phase workflow with clear sequencing, parallelization guidance ('Run these in parallel where possible'), explicit budget constraints ('10-15 tool calls'), quality gates ('Only present opportunities with RICE score >= 100'), and validation checkpoints (cross-referencing across sources, verifying fixes already shipped). The troubleshooting section provides error recovery paths for common failure modes. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear headers and phases, but it's a monolithic ~300-line document with no bundle files to offload detail into. The RICE scoring tables, AI agent analytics field reference, and opportunity template could be split into separate reference files. For a skill of this complexity, the single-file approach makes it harder to navigate. | 2 / 3 |
Total | 10 / 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.
- Repository
- amplitude/builder-skills
- Commit
221ffaa
- 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.