analyzing-text-with-nlp

Execute this skill enables AI assistant to perform natural language processing and text analysis using the nlp-text-analyzer plugin. it should be used when the user requests analysis of text, including sentiment analysis, keyword extraction, topic modeling, or ... Use when analyzing code or data. Trigger with phrases like 'analyze', 'review', or 'examine'.

Install with Tessl CLI

npx tessl i github:jeremylongshore/claude-code-plugins-plus-skills --skill analyzing-text-with-nlp

What are skills?

1.13x

Quality

24%

Does it follow best practices?

Impact

91%

1.13x

Average score across 12 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./plugins/ai-ml/nlp-text-analyzer/skills/analyzing-text-with-nlp/SKILL.md

SKILL.md

Review

Evals

Evaluation results

89%

Mobile App Review Triage

Sentiment analysis with confidence scoring

Criteria

Without context

With context

Sentiment label present

100%

Confidence score present

60%

100%

Confidence is numeric

100%

Confidence for all sentiments

100%

Result links text to label

100%

Structured output format

100%

All reviews analyzed

100%

Standard sentiment categories

25%

62%

Output written to file

100%

No duplicate confidence field

71%

100%

Without context: $0.4118 · 2m 17s · 18 turns · 24 in / 8,003 out tokens

With context: $0.5706 · 2m 21s · 27 turns · 572 in / 7,909 out tokens

100%

Editorial Keyword Intelligence for SEO Planning

Keyword extraction output format

Criteria

Without context

With context

Keywords as list

100%

Per-document keyword sets

100%

Content-specific terms

66%

100%

No stopwords in results

75%

100%

Minimum keyword count

100%

Multi-word phrases included

100%

Structured output format

100%

Output written to file

100%

Without context: $0.2243 · 1m 4s · 15 turns · 22 in / 3,377 out tokens

With context: $0.4783 · 2m 8s · 23 turns · 647 in / 6,496 out tokens

100%

Automated Support Ticket Routing System

Topic modeling with downstream integration

Criteria

Without context

With context

Topics identified

100%

Distinct topic count

100%

Ticket-to-topic assignment

100%

Topic labels are descriptive

100%

Routing rule per topic

100%

Downstream integration output

100%

Structured output format

100%

Output written to file

100%

Without context: $0.2973 · 1m 11s · 14 turns · 19 in / 5,191 out tokens

With context: $0.6329 · 2m 24s · 28 turns · 380 in / 8,584 out tokens

100%

Content Archive Tagging System

Keyword-driven content tagging and organization

Criteria

Without context

With context

Per-article keyword lists

100%

Keywords are specific terms

90%

100%

Keyword list format

100%

Tag taxonomy created

100%

Tags grouped into categories

100%

Article-to-tag mapping

100%

All articles mapped

100%

Structured output format

100%

Output written to file

100%

Without context: $0.2383 · 58s · 10 turns · 13 in / 4,207 out tokens

With context: $1.0938 · 3m 57s · 42 turns · 4,248 in / 13,025 out tokens

100%

Product Feedback Intelligence Brief

Combined multi-type NLP analysis in a single workflow

Criteria

Without context

With context

Sentiment label present

100%

Confidence score present

70%

100%

Keywords as lists

100%

Per-excerpt keyword sets

100%

Topics identified

100%

All three analysis types present

100%

All excerpts analyzed

100%

Content-specific keywords

100%

Unified report output

100%

Structured output format

100%

Output written to file

100%

Without context: $0.3331 · 2m 8s · 11 turns · 16 in / 7,786 out tokens

With context: $0.7507 · 3m 34s · 27 turns · 3,163 in / 11,978 out tokens

90%

14%

Workplace Culture Survey Analysis

Context-enriched analysis with iterative refinement

Criteria

Without context

With context

Domain context noted

100%

Context affects interpretation

100%

Methodology log present

100%

Refinement step documented

26%

53%

Refinement reason stated

70%

Sentiment labels present

100%

Confidence scores present

62%

100%

Keywords extracted

100%

Ambiguous responses flagged

100%

Output written to files

100%

Without context: $0.3309 · 2m 33s · 11 turns · 15 in / 8,121 out tokens

With context: $1.0474 · 4m 56s · 35 turns · 1,676 in / 15,693 out tokens

86%

16%

App Store Review Triage for Product Roadmap

Sentiment-based customer feedback categorization

Criteria

Without context

With context

Sentiment label per review

100%

Confidence score per review

41%

Confidence is numeric

62%

Feedback categorized into action buckets

100%

Category derives from sentiment

75%

100%

All reviews categorized

100%

Structured output format

100%

Output written to file

100%

Integration artifact produced

100%

Confidence differentiates decisions

42%

Without context: $0.3413 · 1m 53s · 14 turns · 21 in / 6,500 out tokens

With context: $0.8175 · 3m 48s · 31 turns · 1,379 in / 12,132 out tokens

93%

100%

14%

Earnings Call Transcript Analysis for Investment Research

Specificity and clarity in multi-type analysis

Criteria

Without context

With context

Analysis plan documents specificity

100%

Domain context stated

100%

Context influences interpretation

100%

Sentiment labels present

100%

Confidence scores present

100%

Keywords as lists

75%

100%

Topics identified

100%

All three analysis types performed

83%

100%

Analysis plan written to file

100%

Results written to file

100%

Without context: $0.3308 · 1m 51s · 15 turns · 20 in / 5,996 out tokens

With context: $0.6633 · 2m 16s · 28 turns · 5,341 in / 7,444 out tokens

35%

Open-Ended Survey Analysis for a SaaS Product Team

Iterative NLP analysis refinement workflow

Criteria

Without context

With context

Multiple analysis passes

26%

Refinement documented

Context in refined pass

Specific refined query

Both passes in output

Sentiment with direction

100%

Keyword list present

100%

Topics identified

100%

Consolidated summary

100%

70%

Insight from refinement

Without context: $0.1766 · 1m 25s · 8 turns · 13 in / 3,927 out tokens

With context: $0.5391 · 2m 25s · 23 turns · 4,122 in / 7,073 out tokens

99%

Content Gap Analysis from Competitor Blog

NLP integration with downstream research actions

Criteria

Without context

With context

Keywords extracted

100%

Keywords drive downstream step

93%

100%

Integration documented

100%

Topic identification present

100%

Sentiment captured

100%

Five or more blog ideas

100%

Ideas tied to analysis

100%

Context enrichment

100%

87%

Two-section structure

100%

Gap framing

100%

Without context: $0.1895 · 1m 21s · 9 turns · 14 in / 4,064 out tokens

With context: $0.4100 · 2m 19s · 16 turns · 1,979 in / 6,577 out tokens

100%

54%

Codebase Health Audit for a Data Pipeline Project

Directory analysis script with CLI output flags

Criteria

Without context

With context

Script invoked

100%

Target directory argument

100%

JSON flag used

100%

Output file flag used

100%

Report file present

50%

100%

Valid JSON report

100%

Statistics referenced

100%

Issues reviewed

100%

Recommendations reviewed

100%

Summary file created

100%

Without context: $0.3046 · 1m 24s · 21 turns · 17 in / 5,974 out tokens

With context: $0.5896 · 1m 41s · 32 turns · 1,997 in / 5,533 out tokens

Evaluated: about 11 hours ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Mobile App Review Triage Editorial Keyword Intelligence for SEO Planning Automated Support Ticket Routing System Content Archive Tagging System Product Feedback Intelligence Brief Workplace Culture Survey Analysis App Store Review Triage for Product Roadmap Related Article Recommendation Engine for a Tech Blog Earnings Call Transcript Analysis for Investment Research Open-Ended Survey Analysis for a SaaS Product Team Content Gap Analysis from Competitor Blog Codebase Health Audit for a Data Pipeline Project

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.