CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dsh-code-review

Use when reviewing a pull request in the deepseek-harness repo — orients the reviewer to this codebase's standards (AGENTS.md conventions, defensive patterns, ADRs, quality gates) and the review-specific checks that code alone can't show

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is dsh-code-review in river666/deepseek-harness-enterprise

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.

An exemplary lean, repo-specific review skill: every line carries non-generic guidance, references are cleanly organized, and the workflow has explicit ordering and retarget recovery. The only improvement space is in actionability and workflow presentation — consolidating the procedure and adding example findings would sharpen both.

DimensionReasoningScore

Conciseness

The body is dense and lean with zero filler — no concept explanations Claude already knows — and compressed phrasing like "Invariant companions are semantic" fully assumes competence. The only redundancy (packages/AGENTS.md linked five times) is point-of-use convenience, not padding, so anchor 5 fits better than 4.

5 / 5

Actionability

It provides a copy-paste command ("pnpm --silent run change-scope --base <verified-base-ref> --head <verified-head-ref>") and specific directives ("trace both sides of every changed interface", "follow every denial path", "probe tiny and exact limits, oversized single chunks, and multibyte text"), but several checks stay at judgment level without examples of what a finding looks like, fitting anchor 4 rather than 5.

4 / 5

Workflow Clarity

The intro states an explicit order ("Verify and fetch the PR's live base and exact head, then run change-scope ... before reading the diff") with feedback loops ("Re-establish the base and rerun it after a retarget or merge") and a required-evidence check, but the sequence and checkpoints are distributed across sections rather than one visible ordered procedure — anchor 4, not 5.

4 / 5

Progressive Disclosure

No bundle files exist, and the ~45-line body is appropriately self-contained: all external references are repo docs collected one level deep under a clearly labeled "Sources of truth" section with well-organized sections throughout, which fits the anchor-5 / simple-skill pattern.

5 / 5

Total

18

/

20

Passed

Description

87%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, tightly scoped description: explicit when-clause, concrete named standards, and unambiguous repo targeting. The only gaps are minor — a slightly meta headline verb and missing synonyms like "PR"/"code review".

DimensionReasoningScore

Specificity

"orients the reviewer to this codebase's standards (AGENTS.md conventions, defensive patterns, ADRs, quality gates)" lists several concrete standards items, though the headline verb ("orients the reviewer") is more abstract than multiple concrete operations, fitting anchor 4 rather than 5.

4 / 5

Completeness

It explicitly answers both parts: the what ("orients the reviewer to this codebase's standards ... and the review-specific checks") and a concrete when-clause ("Use when reviewing a pull request in the deepseek-harness repo"), matching the anchor-5 pattern of explicit what and when with concrete trigger phrases.

5 / 5

Trigger Term Quality

"reviewing a pull request", "codebase's standards", "defensive patterns", "ADRs", "quality gates" are natural terms a user would say, but common synonyms like "PR" and "code review" are missing, so coverage is good rather than comprehensive.

4 / 5

Distinctiveness Conflict Risk

The description is scoped to one named repository and one task (PR review), giving it a clear niche with distinct triggers and minimal risk of firing for the wrong skill.

5 / 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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 19 suspicious

Warning

Total

15

/

16

Passed

Repository

river666/deepseek-harness-enterprise

Path

.claude/skills/dsh-code-review/SKILL.md

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.