opencode-skills-collection 4.0.53 → 4.0.55

package/bundled-skills/selenium-skill/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: selenium-skill
-description: Generates production-grade Selenium WebDriver automation scripts and tests in Java, Python, JavaScript, C#, Ruby, or PHP. Supports local execution and TestMu AI cloud with 3000+ browser/OS combinations. Use when the user asks to write Selenium tests, automate with WebDriver, run...
+description: "Generates production-grade Selenium WebDriver automation scripts and tests in Java, Python, JavaScript, C#, Ruby, or PHP. Supports local execution and TestMu AI cloud with 3000+ browser/OS combinations."
 risk: critical
 source: https://github.com/LambdaTest/agent-skills/tree/main/selenium-skill
 source_repo: LambdaTest/agent-skills

package/bundled-skills/semgrep-rule-creator/SKILL.md

@@ -11,6 +11,7 @@ allowed-tools:
   - WebFetch
 risk: critical
 source: community
+date_added: "2026-09-04"
 ---
 
 # Semgrep Rule Creator

package/bundled-skills/semgrep-rule-variant-creator/SKILL.md

@@ -5,6 +5,7 @@ allowed-tools:
  ...
 risk: critical
 source: community
+date_added: "2026-09-04"
 ---
 
 # Semgrep Rule Variant Creator

package/bundled-skills/sendgrid-automation/SKILL.md

@@ -230,6 +230,12 @@ Contact operations (`ADD_OR_UPDATE_A_CONTACT`, `IMPORT_CONTACTS`) are asynchrono
 ## When to Use
 This skill is applicable to execute the workflow or actions described in the overview.
 
+## Example
+
+**User request:**
+
+> Automate SendGrid email delivery workflows including marketing campaigns (Single Sends), contact and list management, sender identity setup, and email analytics through Composio's SendGrid toolkit.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-audit/SKILL.md

@@ -112,9 +112,11 @@ If critical context is missing, **state assumptions explicitly** before proceedi
 
 **Key Metrics**
 
-* LCP < 2.5s
-* INP < 200ms
-* CLS < 0.1
+* LCP at or below 2.5s
+* INP at or below 200ms
+* CLS at or below 0.1
+
+Evaluate field data at the 75th percentile with the relevant device population. Keep lab diagnostics distinct from observed field performance; see [Google’s Core Web Vitals guidance](https://developers.google.com/search/docs/appearance/core-web-vitals).
 
 **Contributing Factors**
 
@@ -218,7 +220,7 @@ If critical context is missing, **state assumptions explicitly** before proceedi
 
 ### Purpose
 
-The **SEO Health Index** provides a **normalized, explainable score** that summarizes overall SEO health **without replacing detailed findings**.
+The optional **SEO Health Index** is a subjective audit rubric, not a validated search-engine metric. It can organize findings only when its weights, scope and deductions are disclosed. Do not invent points for unavailable data or imply an effect on rankings.
 
 It is designed to:
 
@@ -244,7 +246,7 @@ The score is a **weighted composite**, not an average.
 | Authority & Trust Signals | 10      |
 | **Total**                 | **100** |
 
-> If a category is **out of scope**, redistribute its weight proportionally and state this explicitly.
+> Mark uninspected categories as unknown. If presenting a scoped score, name its scope and do not compare it with an earlier full-site score.
 
 ---
 
@@ -267,8 +269,7 @@ Start each category at **100** and subtract points based on issues found.
 
 #### Confidence Modifier
 
-If confidence is **Medium**, apply **50%** of the deduction
-If confidence is **Low**, apply **25%** of the deduction
+Keep confidence separate from severity. An uncertain critical finding needs verification; reducing its numerical deduction does not make the site safer to index.
 
 ---
 
@@ -277,11 +278,11 @@ If confidence is **Low**, apply **25%** of the deduction
 > Crawlability & Indexation (Weight: 30)
 
 * Noindex on key category pages → Critical (−25, High confidence)
-* XML sitemap includes redirected URLs → Medium (−5, Medium confidence → −2.5)
+* XML sitemap includes redirected URLs → Medium (−5, Medium confidence; verify separately)
 * Missing sitemap reference in robots.txt → Low (−2)
 
-**Raw score:** 100 − 29.5 = **70.5**
-**Weighted contribution:** 70.5 × 0.30 = **21.15**
+**Raw score:** 100 − 32 = **68**
+**Weighted contribution:** 68 × 0.30 = **20.4**
 
 ---
 
@@ -298,9 +299,9 @@ Rounded to nearest whole number.
 
 ---
 
-## Health Bands (Required)
+## Illustrative bands (optional)
 
-Always classify the final score into a band:
+If a rubric is requested, label its bands as internal planning labels rather than measured SEO performance:
 
 | Score Range | Health Status | Interpretation                                  |
 | ----------- | ------------- | ----------------------------------------------- |
@@ -314,7 +315,7 @@ Always classify the final score into a band:
 
 ## Output Requirements (Scoring Section)
 
-Include this **after the Executive Summary**:
+If a score is requested, include this after the evidence-based findings:
 
 ### SEO Health Index
 
@@ -398,7 +399,7 @@ These fields are **mandatory** and directly inform the SEO Health Index.
   A short explanation of the SEO impact in plain language.
 
 * **Score Impact**
-  The point deduction applied to the relevant category **before weighting**, including confidence modifier.
+  Optional rubric deduction, with confidence reported separately; omit when no score is requested.
 
 * **Recommendation**
   What should be done to resolve the issue.
@@ -408,7 +409,7 @@ These fields are **mandatory** and directly inform the SEO Health Index.
 
 ### Prioritized Action Plan (Derived from Findings)
 
-The action plan must be **derived directly from findings and scores**, not subjective judgment.
+Derive the action plan from observed findings, affected URLs and product importance. A subjective score must not override crawl/indexation evidence.
 
 Group actions as follows:
 
@@ -426,7 +427,7 @@ Group actions as follows:
 3. **Quick Wins**
 
    * Low or Medium severity issues
-   * Easy to fix with measurable score improvement
+   * Easy to fix with a verifiable user or crawlability benefit
 
 4. **Longer-Term Opportunities**
 
@@ -436,35 +437,16 @@ Group actions as follows:
 For each action group:
 
 * Reference the **related findings**
-* Explain **expected score recovery range**
+* Define the observable verification after remediation
 * Avoid timelines unless explicitly requested
 
 ---
 
-### Tools (Evidence Sources Only)
-
-Tools may be referenced **only to support evidence**, never as authority by themselves.
-
-Acceptable uses:
-
-* Demonstrating an issue exists
-* Quantifying impact
-* Providing reproducible data
-
-Examples:
-
-* Search Console (coverage, CWV, indexing)
-* PageSpeed Insights (field vs lab metrics)
-* Crawlers (URL discovery, metadata validation)
-* Log analysis (crawl behavior, frequency)
-
-Rules:
+### Evidence sources
 
-* Do not rely on a single tool for conclusions
-* Do not report tool “scores” without interpretation
-* Always explain *what the data shows* and *why it matters*
-
----
+Use Search Console for reported indexing and field data, crawlers for sampled URL
+behavior, and logs for observed requests. Record dates, scope and missing coverage.
+Cross-check material claims against the actual page; tool scores are not authority.
 
 ### Related Skills (Non-Overlapping)
 
@@ -484,9 +466,16 @@ Use these skills **only after the audit is complete** and findings are accepted.
 
 
 ## When to Use
-This skill is applicable to execute the workflow or actions described in the overview.
+
+Use for a specific indexing/crawlability issue, a migration check or a scoped organic-performance audit. Start from the declared production URLs and available Search Console/crawl evidence; local source changes do not prove live deployment or indexing.
+
+## Worked example
+
+Input: a legacy URL redirects to a new catalog. Fetch the legacy URL and every redirect hop, inspect final status/canonical/robots directives, and compare the destination with sitemap/internal links. Expected: a finite redirect chain to the intended indexable URL, with no conflicting canonical. Then check Search Console separately; a correct HTTP response does not prove that Google selected or indexed it.
 
 ## Limitations
-- Use this skill only when the task clearly matches the scope described above.
-- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
-- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.
+
+- Rankings and traffic depend on demand, competition and search-engine processing; technical corrections do not guarantee recovery.
+- A sampled crawl cannot establish that every URL or rendered state is correct.
+- Canonical tags are signals, and sitemap inclusion is not proof of indexation.
+- Lab performance, field percentiles and an internal audit score answer different questions; report the source and date of each.

package/bundled-skills/seo-content/SKILL.md

@@ -184,6 +184,12 @@ If DataForSEO MCP tools are available, use `kw_data_google_ads_search_volume` fo
 | Content behind paywall (402/403, login wall) | Report that the content is not publicly accessible. Analyze only the visible portion (meta tags, headers) and note the limitation. |
 | Thin content (fewer than 100 words retrievable) | Report the findings as-is rather than guessing. Flag the page as potentially JavaScript-rendered or gated, and suggest the user provide the full text directly. |
 
+## Example
+
+**User request:**
+
+> Audit content quality, readability, thin content risk, or E-E-A-T signals.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-content-auditor/SKILL.md

@@ -81,6 +81,12 @@ You are an SEO content auditor analyzing provided content for optimization oppor
 
 Focus on actionable improvements based on SEO best practices and content quality standards.
 
+## Example
+
+**User request:**
+
+> Analyze the supplied material with @seo-content-auditor, identify the highest-priority findings, and give concrete next steps.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-content-planner/resources/implementation-playbook.md

@@ -0,0 +1,23 @@
+# Content planning playbook
+
+## Inputs
+
+Product/audience brief, existing pages, target market and supplied or authorized search evidence.
+
+## Procedure
+
+1. Group questions by user intent and identify existing pages that already answer them. Separate observed search evidence from topic hypotheses.
+2. Create an outline with the reader's question, useful answer, required sources and next action. Assign one primary role per page and explicit internal links.
+3. Prioritize by audience relevance, evidence and editorial capacity. Deliver a calendar with owners and dependencies rather than invented traffic estimates.
+
+## Worked example
+
+A small software site needs onboarding articles. Map installation, first task and troubleshooting into distinct pages, reusing existing documentation and flagging missing product evidence.
+
+## Verification and handoff
+
+Report the actual files or configuration changed, checks performed, observed results and any untested environment. Keep the original inputs and evidence sufficient to reproduce the conclusion.
+
+## Limitations
+
+Do not invent search volume, rankings or customer research. Word count is an editorial constraint, not a ranking target.

package/bundled-skills/seo-content-writer/SKILL.md

@@ -79,14 +79,14 @@ You are an SEO content writer creating comprehensive, engaging content optimized
 
 **Quality Standards:**
 - Original, valuable content
-- 0.5-1.5% keyword density
+- Natural keyword use without a fixed density target
 - Grade 8-10 reading level
 - Short paragraphs (2-3 sentences)
 - Bullet points for scannability
 - Examples and data support
 
 **E-E-A-T Elements:**
-- First-hand experience mentions
+- First-hand experience only when supplied and verifiable
 - Specific examples and cases
 - Data and statistics citations
 - Expert perspective inclusion
@@ -94,6 +94,12 @@ You are an SEO content writer creating comprehensive, engaging content optimized
 
 Focus on value-first content. Write for humans while optimizing for search engines.
 
+## Example
+
+**User request:**
+
+> Review this campaign asset with @seo-content-writer, identify the main conversion problem, and produce an improved version grounded in the available evidence.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-content-writer/resources/implementation-playbook.md

@@ -0,0 +1,23 @@
+# Evidence-grounded content drafting
+
+## Inputs
+
+Approved brief, reader intent, factual sources, product claims and editorial voice.
+
+## Procedure
+
+1. Outline the answer before adding search terms. Keep sourced facts, supplied experience and illustrative examples distinguishable.
+2. Write clear sections that answer the reader's question. Use keywords naturally; do not force density targets or fabricate first-hand experience, quotations or statistics.
+3. Check every factual claim and proposed internal link. Provide title and metadata suggestions, noting unresolved facts and any language requiring specialist review.
+
+## Worked example
+
+Draft a setup guide from verified product documentation. Include steps, expected outcome and troubleshooting; label an untested step instead of claiming it worked.
+
+## Verification and handoff
+
+Report the actual files or configuration changed, checks performed, observed results and any untested environment. Keep the original inputs and evidence sufficient to reproduce the conclusion.
+
+## Limitations
+
+A polished draft does not imply ranking gains or publication. Do not manufacture expertise signals or publish without authorization.

package/bundled-skills/seo-forensic-incident-response/SKILL.md

@@ -260,6 +260,12 @@ When helping a user, ask:
 - **schema-markup**: For implementing structured data after stability is restored.
 - **analytics-tracking**: For ensuring measurement is correct post-incident.
 
+## Example
+
+**User request:**
+
+> Investigate this sudden organic-traffic drop, establish the timeline and likely cause, and produce a prioritized recovery plan.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-fundamentals/SKILL.md

@@ -172,6 +172,12 @@ SEO fundamentals should be validated using **multiple signals**, not single metr
 ## When to Use
 This skill is applicable to execute the workflow or actions described in the overview.
 
+## Example
+
+**User request:**
+
+> Review this campaign asset with @seo-fundamentals, identify the main conversion problem, and produce an improved version grounded in the available evidence.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/seo-programmatic/SKILL.md

@@ -182,6 +182,12 @@ Measure against all other pages in the programmatic set. Shared headers, footers
 | Thin content threshold exceeded | Trigger quality gate warning. Report the unique content percentage and flag pages below 40% uniqueness. Require user acknowledgment before proceeding. |
 | Quality gate violation | Halt analysis at the HARD STOP threshold (500+ pages without justification or <30% unique content). Present findings and require explicit user approval to continue. |
 
+## Example
+
+**User request:**
+
+> Plan and audit programmatic SEO pages generated at scale from structured data.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/sequence-psychologist/SKILL.md

@@ -113,6 +113,12 @@ Before finalizing output, the agent asks:
 - [ ] Did I resolve open loops on time?
 - [ ] Does the sequence feel respectful and useful?
 
+## Example
+
+**User request:**
+
+> Rewrite this email sequence with a clearer step-by-step persuasion arc and credible transitions.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/server-management/SKILL.md

@@ -165,6 +165,12 @@ When something's wrong:
 ## When to Use
 This skill is applicable to execute the workflow or actions described in the overview.
 
+## Example
+
+**User request:**
+
+> Use @server-management for this task: Server management principles and decision-making.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/service-mesh-expert/SKILL.md

@@ -60,6 +60,12 @@ Expert service mesh architect specializing in Istio, Linkerd, and cloud-native n
 - Keep sidecar resources appropriately sized
 - Use destination rules for consistent load balancing
 
+## Example
+
+**User request:**
+
+> Implement service-to-service communication in Kubernetes.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/setup-help/SKILL.md

@@ -44,6 +44,12 @@ Repeat this format for every response until setup is done.
 - Update the "Still remaining" list each time as steps get done.
 - When nothing remains, say setup is complete instead of showing the list.
 
+## Example
+
+**User request:**
+
+> Use @setup-help for this task: Walk a user through setup or installation one step at a time with the remaining steps visible.
+
 ## Limitations
 
 - Adapted from `davidondrej/skills`; verify local paths, tools, credentials, and agent features before acting.