Content
78%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A highly actionable, well-sequenced setup guide with executable SQL at every step and explicit end-to-end verification. Its weaknesses are mild redundancy and general-Postgres filler, plus a monolithic single-file layout with no progressive disclosure into reference files despite being long enough to warrant it.
Suggestions
Split stable reference material (deprecated-vs-current API mapping, schema best-practices checklist, sparse index reference) into files under references/ and link them from short SKILL.md sections, moving the body toward an overview-plus-detail structure.
Trim the Do's and Don'ts list to the TimescaleDB-specific items (TIMESTAMPTZ, time_bucket usage) and drop general PostgreSQL conventions Claude already knows, and de-duplicate the chunk-size guidance repeated in both the chunk-interval section and Performance Guidelines.
Add a short error-recovery step after Step 11 verification: what to check and fix when chunks remain uncompressed, jobs fail, or aggregate refresh falls behind, turning the verify queries into a validate→fix→retry loop.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with genuinely non-obvious, domain-specific guidance (row-density thresholds, tsdb.* DDL options, deprecated→current API mapping) and avoids explaining basics, so it is mostly efficient. However, the Do's and Don'ts list reiterates general PostgreSQL practices Claude already knows ("Use snake_case NOT CamelCase", "Use NOT EXISTS NOT NOT IN"), and some guidance (chunk-size test, retention caution) is repeated across sections — minor over-explanation that could be trimmed, fitting the level-4 anchor rather than the lean level-5 anchor. | 4 / 5 |
Actionability | Every step ships copy-paste-ready SQL with placeholders: the CREATE TABLE ... WITH (tsdb.hypertable, ...) template, ALTER TABLE columnstore settings, add_retention_policy/add_continuous_aggregate_policy calls, index DDL, and verification queries. The examples cover the common cases (IoT, finance, metrics, event logs) with concrete parameter values, matching the fully-executable top anchor. | 5 / 5 |
Workflow Clarity | Steps 1–11 form a clear, well-ordered sequence, and Step 11 provides explicit verification queries (hypertables, compression stats, jobs, chunks) — so the destructive/batch-operation validation cap at 3 does not apply. It falls short of the level-5 anchor because there is no error-recovery feedback loop (e.g., what to do when verify queries show uncompressed chunks or a misconfigured policy) and some checkpoints (e.g., the segment_by 'good test') are stated but not tied back into a validation step. | 4 / 5 |
Progressive Disclosure | No bundle files exist (references/, scripts/, assets/ are absent) and the ~475-line body keeps everything inline — including an API reference section and schema best-practices list that would naturally live in referenced files. Section headers do give reasonable in-file organization, but there is no overview-to-detail split across files, fitting 'some structure but could be better organized; content that should be separate is inline'. | 3 / 5 |
Total | 16 / 20 Passed |