dt-obs-tracing
Distributed traces, spans, service dependencies, and request flow analysis. Use when investigating span-level details, failures, performance bottlenecks, or trace correlation. Trigger: "trace analysis", "slow requests", "failed spans", "service dependencies", "distributed trace", "span details", "HTTP status codes in traces", "database query spans", "messaging spans", "gRPC calls", "Lambda cold starts", "trace ID lookup", "exception analysis", "correlate logs and traces", "request attributes". Do NOT use for explaining existing queries, product documentation or configuration questions, service-level RED metrics (use dt-obs-services), log searching (use dt-obs-logs), or problem analysis (use dt-obs-problems).
71
86%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
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 covers all dimensions thoroughly. It provides specific capabilities, extensive natural trigger terms, clear 'Use when' and 'Do NOT use' guidance, and explicit boundaries with related skills. The negative triggers referencing other specific skills are particularly effective for disambiguation in a multi-skill environment.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: distributed traces, spans, service dependencies, request flow analysis, span-level details, failures, performance bottlenecks, trace correlation. Very detailed about what it covers. | 3 / 3 |
Completeness | Clearly answers both 'what' (distributed traces, spans, service dependencies, request flow analysis) and 'when' (explicit 'Use when' clause plus extensive trigger list). Also includes explicit 'Do NOT use' guidance with redirects to other skills, which further strengthens completeness. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'slow requests', 'failed spans', 'trace ID lookup', 'database query spans', 'Lambda cold starts', 'HTTP status codes in traces', 'correlate logs and traces'. These are highly natural phrases a user would use when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with clear boundaries. The 'Do NOT use' section explicitly delineates this skill from related skills (dt-obs-services, dt-obs-logs, dt-obs-problems), making it very unlikely to conflict with similar observability skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
72%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, comprehensive skill for distributed trace analysis with excellent actionability through numerous executable DQL queries and outstanding progressive disclosure via well-organized references. The main weaknesses are moderate verbosity in conceptual sections (span kinds, use cases with trigger words that duplicate frontmatter metadata) and a lack of explicit multi-step investigation workflows with validation checkpoints — queries are presented as isolated patterns rather than guided diagnostic sequences.
Suggestions
Remove or significantly trim the 'Use Cases' section (trigger words belong in frontmatter) and the 'Core Concepts' explanation of what spans are — Claude knows these concepts. Keep only the Dynatrace-specific attribute table and sampling details.
Add at least one explicit multi-step investigation workflow (e.g., 'Diagnosing a slow endpoint: 1. Run percentile query → 2. If p99 > threshold, find exemplar traces → 3. Look up trace spans → 4. Identify bottleneck span → 5. Check for database/external call issues') with decision points and verification steps.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is generally well-structured but includes some unnecessary content like the 'Use Cases' section with trigger words (metadata that belongs in frontmatter), the 'Core Concepts' section explaining what spans and span kinds are (Claude likely knows this), and the attribute table which is somewhat verbose. However, the query examples themselves are lean and purposeful. | 2 / 3 |
Actionability | Excellent actionability throughout — nearly every section includes complete, executable DQL queries that are copy-paste ready. The queries cover a wide range of real scenarios (slow trace detection, failure analysis, service dependencies, trace aggregation) with specific field names, filter conditions, and output formatting. | 3 / 3 |
Workflow Clarity | While individual queries are clear and well-organized by analysis type, the skill lacks explicit multi-step investigation workflows with validation checkpoints. For example, trace investigation typically involves a sequence (find anomalous traces → drill into spans → correlate with logs), but this is presented as isolated query patterns rather than guided workflows with decision points and verification steps. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure with a clear overview in the main file and well-signaled one-level-deep references to 12 specific reference files covering detailed topics (performance analysis, failure detection, sampling, span types, etc.). The span types table elegantly maps each type to its reference file, and inline '📖 Learn more' links provide contextual navigation. | 3 / 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
- Dynatrace/dynatrace-for-ai
- Commit
7cbe1ef
- 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.