CtrlK
BlogDocsLog inGet started
Tessl Logo

review-performance

Use when reviewing a branch diff for performance defects — redundant database round trips, N+1 queries, unbounded reads, per-request layer rebuilds, wasted client fetches, and build steps that ship a broken artefact — and reporting findings with P-C/W/I IDs in the four-field format.

69

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

A highly actionable, well-sequenced review procedure with genuine validation feedback loops and correct use of a one-level-deep reference file. Its weakness is repetition: the report-shape discipline is restated four times with emphatic padding, and the "rest of the checklist" section miscounts its own list, both of which cost tokens without adding instruction.

Suggestions

Fix the count in "Eight more sections live in references/checklists.md" — ten sections are named and ten exist in the reference file; also make the reference a markdown link ([checklists.md](references/checklists.md)) so it is unambiguous.

Consolidate the shape-discipline guidance: state the four-heading rule and the grep -c check once (Step 0 or Report shape) and have the other three mentions reference it in a single line instead of re-explaining the failure mode each time.

Move the long two-gate bullet under Shared middleware (or its worked two-question walkthrough) into references/checklists.md alongside the other deferred sections, keeping only the summary line and the routing trigger in SKILL.md.

DimensionReasoningScore

Conciseness

The body is operational and repo-specific rather than explanatory, but the skeleton/shape discipline is explained four separate times (Step 0's two long paragraphs, the Step 2 mid-run check, "Report shape", and the final check), with rhetorical padding ("a run that leaves a differently-shaped file has produced nothing, however good the analysis inside it") and a factual slip ("Eight more sections" introduces a list of ten). This fits the mostly-efficient-but-could-be-tightened anchor at 3 rather than the minor-trimmable-instances anchor at 4, and is well above the noticeably-verbose anchor at 2.

3 / 5

Actionability

Guidance is fully executable: a verbatim heredoc that writes the report skeleton, exact base-branch resolution and diff-routing commands, a grep routing table mapping patterns to mandatory sections, a `grep -c` shape check, a copy-paste finding template, and a named in-repo fix idiom (`cire/api/src/services/directory.ts` and its `inWedding` column) to copy rather than invent.

5 / 5

Workflow Clarity

The sequence (write skeleton → route the diff → work mandatory sections → assemble the report) is explicit, with validation checkpoints at three points (mid-run `grep -c` after the first finding, a final two-count check, and the recovery loop "restore the 4 headings, put the finding back under the right one, and edit from then on") and per-bullet verdict recording to catch skimming. This matches the clear-sequence-with-explicit-validation-and-error-recovery anchor.

5 / 5

Progressive Disclosure

Ten checklist sections are routed to a single one-level-deep reference (`references/checklists.md`, verified to exist with matching headings) and clearly signaled ("open the file whenever Step 1 routes to one of them"), with the highest-frequency section kept inline. It falls short of the 5 anchor on minor organization gaps: the miscounted "Eight more sections" heading, no markdown link to the reference file, and the very long two-gate bullet under Shared middleware that could itself live in the reference.

4 / 5

Total

17

/

20

Passed

Description

88%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.

A strong description: it names the domain, enumerates the specific defect classes it covers, states the output format, and opens with an explicit "Use when" trigger clause. The only gaps are a few missing natural synonyms and mild overlap with general code-review skills on the word "reviewing".

DimensionReasoningScore

Specificity

The description enumerates the concrete defect classes it handles ("redundant database round trips, N+1 queries, unbounded reads, per-request layer rebuilds, wasted client fetches, and build steps that ship a broken artefact") and specifies the exact output convention ("P-C/W/I IDs in the four-field format"), matching the comprehensive-coverage anchor rather than the minor-gaps anchor at 4.

5 / 5

Completeness

It opens with an explicit trigger clause ("Use when reviewing a branch diff for performance defects") and pairs it with a concrete what (the named defect classes plus "reporting findings with P-C/W/I IDs in the four-field format"), clearly and explicitly answering both what and when with concrete triggers.

5 / 5

Trigger Term Quality

Natural trigger phrases like "performance defects", "branch diff", "N+1 queries" and "database round trips" are present, but common synonyms a user might say ("slow queries", "latency", "perf") are missing, which fits the good-coverage-with-a-few-gaps anchor rather than the comprehensive-synonyms anchor at 5.

4 / 5

Distinctiveness Conflict Risk

The performance-only scope and named defect classes give it a clear niche, but the trigger phrase "reviewing a branch diff" overlaps with general code-review skills that would also fire on a plain review request, placing it at mostly-distinct with minor overlap risk rather than minimal conflict risk at 5.

4 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
englishstventures/osn
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.