seo-backlinks

Analyze a site's backlink profile, anchors, toxic signals, competitors, gaps, and disavow candidates. Use only when links or referring domains are the requested focus.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

67%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 operational skill body: concrete commands, numeric thresholds, source fallback cascades, and a genuine validate-and-fix feedback loop around scoring integrity. Its weaknesses are redundancy (the no-numeric-score rule stated three ways), a pre-delivery checklist contaminated with items from a different SEO skill, and unspecified input formats for the link-file and validator report.

Suggestions

State the Common Crawl "no numeric score" rule once (e.g., in the data sufficiency gate section) and have the MUST NOT section and validator description reference it, cutting roughly 20 lines of repetition.

Remove or relocate the off-topic Pre-Delivery Review items (parse_html @type checks, h1_suspicious headings) to the on-page SEO skill they belong to, so the backlink checklist only covers link claims.

Specify the expected format of the "--links <file>" input and the validator's "--report <report>.json" structure (a short example JSON snippet), so the verify and validation-gate commands are fully executable end to end.

Dimension	Reasoning	Score
Conciseness	The body is mostly dense, high-value tables and commands with no concept tutorials, but the Common-Crawl "no numeric score" rule is stated three times (data sufficiency gate, "MUST NOT" section, and the validator description), and the Pre-Delivery Review includes off-topic items ("parse_html @type", "h1_suspicious" H1s) from an on-page SEO skill. This is more than the "minor instances" of anchor 4 but well short of the padding of anchor 2.	3 / 5
Actionability	Nearly everything is executable: exact script invocations with flags (e.g. "run moz_api.py metrics <url> --json"), per-source preference orders, numeric thresholds in tables, and an error-to-resolution table. Minor gaps: the format of the "--links <file>" input and the validator's "--report <report>.json" JSON is never specified, and DataForSEO MCP tools are named without parameters.	4 / 5
Workflow Clarity	The sequence is clear (source detection first, then the 7-section analysis, scoring with a data sufficiency gate, validation, output) and includes an explicit feedback loop ("If the validator returns status: FAIL, do not present the report... re-run until it passes") plus a pre-delivery checklist. However, the flow is scattered (Error Handling and the checklist come after Output Format) and the checklist mixes in off-topic on-page items, keeping it below anchor 5.	4 / 5
Progressive Disclosure	References are well-signaled with explicit load conditions ("Load on demand (do NOT load at startup)") and are one level deep, matching anchor 4. It falls short of anchor 5 because the two referenced files are not part of this bundle, and the same reference is cited through two inconsistent paths ("../seo/references/backlink-quality.md" vs "${CLAUDE_PLUGIN_ROOT}/skills/seo/references/backlink-quality.md").	4 / 5
	Total	15 / 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 description: it states concrete capabilities with domain synonyms and an explicit, scope-restricting trigger clause that clearly separates it from sibling SEO skills. The only weakness is that capabilities are compressed into a single "Analyze" verb with a list of objects rather than distinct action verbs, leaving a few natural trigger phrasings implicit.

Dimension	Reasoning	Score
Specificity	"Analyze a site's backlink profile, anchors, toxic signals, competitors, gaps, and disavow candidates" enumerates six concrete analysis targets, but all are objects of the single verb "Analyze" rather than several distinct actions as in the anchor-5 example ("Extract... fill... merge... convert..."). It is clearly above anchor 3, which covers only 1-2 concrete actions.	4 / 5
Completeness	The "what" is explicit (enumerated backlink analysis targets) and the "when" is an explicit trigger clause ("Use only when links or referring domains are the requested focus") with concrete trigger phrases, matching the anchor-5 example structure of statement plus "Use when..." guidance.	5 / 5
Trigger Term Quality	Natural domain terms are present with synonyms ("backlink profile", "anchors", "toxic signals", "disavow candidates", "links or referring domains"), but some plain user phrasings ("backlinks", "link building", "spammy links") are only partially covered by jargon like "toxic signals" and "gaps". Falls between anchors 4 and 5.	4 / 5
Distinctiveness Conflict Risk	Off-site link analysis is a clear niche with distinct triggers, and the scope-restricting "Use only when..." clause minimizes conflict with sibling SEO skills (on-page audits, keyword research). Not anchor 4, which implies minor overlap risk with closely related skills.	5 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: AgriciDaniel/claude-seo
Path: skills/seo-backlinks/SKILL.md
Commit: 4b99de2

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.