CtrlK
BlogDocsLog inGet started
Tessl Logo

dt-dql-essentials

Core DQL syntax, pitfalls, query patterns, and query optimization. Load to write, build, fix, or OPTIMIZE a DQL query — prevents syntax errors and makes queries faster, more efficient, and cheaper (less data scanned = lower query consumption/cost per run). Covers fetch commands, data models, field namespaces, time alignment, entity/smartscape patterns, metric discovery, and performance/cost optimization (filter early, bucket filters, short time ranges, field selection, sampling, cardinality). Trigger: "write/build/fix a DQL query", "DQL syntax", "query logs/spans/metrics", "create a timeseries", "optimize my DQL", "make my query faster/cheaper", "reduce DQL cost/consumption/scanned data", "keep DQL cost under control". Do NOT use to explain an existing query or answer product questions. For MONITORING a tenant's ACTUAL query consumption/billing (how much queries cost, who scanned most, cost trends) use dt-platform-costs — this tunes the query text, not billing data.

93

1.54x
Quality

93%

Does it follow best practices?

Impact

94%

1.54x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a strong, highly actionable DQL reference: executable snippets, a comprehensive wrong→right pitfall table, and clean one-level-deep routing to real bundle files. Its main defects are a verbatim duplicated 'makeTimeseries' section and repeated rollup:/optimization caveats, plus the absence of an explicit ordered query-authoring workflow.

Suggestions

Delete the second 'makeTimeseries Command' section (lines 357–382) and merge its unique content (the `spread:` entity-timeline example and the `{}`-grouped aggregation form) into the first occurrence to remove the duplication.

Add a short ordered workflow for authoring a DQL query (e.g., discover fields/data objects via `describe` or references/discovery.md → write the query → check the Syntax Pitfalls table → apply references/optimization.md techniques) so the sections read as a sequence rather than a catalog.

Consolidate the rollup:/timeseries-only rule, which is stated three times (pitfalls table, Timeseries Aggregation Functions, and 'The rollup: parameter'), into one authoritative section with pointers to it.

DimensionReasoningScore

Conciseness

The body is dense with DQL-specific, non-obvious knowledge (37-row pitfall table with wrong/right pairs, fetch→data-model mapping) and does not explain concepts Claude already knows, but there is real redundancy: the "makeTimeseries Command" section appears twice (lines 278–297 and 357–382), and the rollup:/timeseries-only caveat is repeated in the pitfalls table, the aggregation table, and its own subsection. This fits 'Efficient; minor instances of over-explanation that could be trimmed' better than the lean 5 anchor. It is not a 3 because the padding is duplication rather than unnecessary explanation.

4 / 5

Actionability

Nearly everything is executable: copy-paste-ready DQL snippets (samplingRatio extrapolation query, chained lookup patterns, timeseries scalar:true forms), a wrong→right pitfall table with exact error names (UNKNOWN_PARAMETER_DEFINED, FIELD_DOES_NOT_EXIST), and concrete discovery commands like `fetch dt.system.data_objects | fields name, display_name, type`. This matches 'Fully executable; copy-paste ready code or commands; specific examples cover the common cases'. Not a 4 because guidance goes beyond code to precise failure modes and fixes for common cases.

5 / 5

Workflow Clarity

The 'When to Load References' table gives a clear task→reading routing, the pitfall table acts as an error-prevention checklist, and verification cues exist (e.g., `describe dt.entity.<type>` before selecting fields, checking `dt.system.sampling_ratio`). However, there is no explicit ordered sequence for authoring a query (discover → write → check pitfalls → optimize), and the duplicated makeTimeseries section weakens coherence. Fits 'Clear sequence with most checkpoints present; minor validation gaps'. Not a 3 because routing and checkpoints are present and this is a read-only query skill where destructive-operation validation caps do not apply.

4 / 5

Progressive Disclosure

The body is an overview with well-signaled, one-level-deep references: a task→reference table, a function-group→spec index for references/dql/*.md, and inline pointers after each section. All referenced files exist in the bundle (references/*.md and references/dql/*.md), and no reference chains deeper than one level. This matches 'Clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation'. Not a 4 because both routing tables are complete and every pointer resolves to a real file.

5 / 5

Total

18

/

20

Passed

Description

100%Weight 40%Scale 1-5

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 is explicit, concrete, and highly triggerable: it states what the skill covers, lists natural user trigger phrases, defines negative triggers, and routes billing/monitoring requests to the sibling dt-platform-costs skill. Its only weakness is verbosity — the optimization coverage list is stated twice and parentheticals pad the text — but every claim is specific and earns its place for retrieval purposes.

DimensionReasoningScore

Specificity

The description lists multiple concrete capabilities — "write, build, fix, or OPTIMIZE a DQL query", "prevents syntax errors", "makes queries faster, more efficient, and cheaper" — and enumerates coverage areas ("fetch commands, data models, field namespaces, time alignment, entity/smartscape patterns, metric discovery, and performance/cost optimization"). Actions are concrete and comprehensive rather than generic, matching the anchor 'Lists multiple specific concrete actions; comprehensive coverage'. It is not a 4 because coverage of the domain's tasks is thorough, not just 'several actions with minor gaps'.

5 / 5

Completeness

It explicitly answers 'what' ("Core DQL syntax, pitfalls, query patterns, and query optimization") and 'when' (an explicit "Trigger:" clause with concrete phrases), plus 'when not' ("Do NOT use to explain an existing query or answer product questions"). This matches the anchor 'Clearly and explicitly answers both what AND when with concrete trigger phrases'. It is not a 4 because the 'when' is fully explicit rather than merely present.

5 / 5

Trigger Term Quality

The trigger list gives natural user phrasings with variations: "write/build/fix a DQL query", "DQL syntax", "query logs/spans/metrics", "create a timeseries", "optimize my DQL", "make my query faster/cheaper", "reduce DQL cost/consumption/scanned data", "keep DQL cost under control". These are the exact phrases a user needing this skill would say, covering syntax, authoring, and optimization intents with synonyms. Not a 4 because no common variation of the intent is obviously missing.

5 / 5

Distinctiveness Conflict Risk

The DQL-query niche is distinct, and the description actively disambiguates from the nearest sibling skill: "For MONITORING a tenant's ACTUAL query consumption/billing ... use dt-platform-costs — this tunes the query text, not billing data". With explicit negative triggers and cross-skill routing, conflict risk is minimal, matching the 'clear niche with distinct triggers' anchor. Not a 4 because overlap risk is addressed rather than merely minor.

5 / 5

Total

20

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 28 deeper-than-1-level

Warning

referenced_paths_exist

Referenced path issues: 31 deeper-than-1-level

Warning

Total

14

/

16

Passed

Repository
Dynatrace/dynatrace-for-ai
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.