perplexity-web-research

Name: perplexity-web-research
Author: xpepper

96installs2forks

How to install perplexity-web-research?

xpepper/perplexity-agent-skill/perplexity-web-research

$npx -y skills add xpepper/perplexity-agent-skill --skill perplexity-web-research

Installs into the current project.

Summary

Leverage Perplexity CLI for web research, deep reasoning, and independent validation. Use when you need current/external knowledge, want a second opinion before decisions, or want to delegate complex analytical thinking to specialized reasoning models. Complements (not replaces)

Files · 1

View on GitHub

SKILL.md

1	# Perplexity Web Research
2
3	Use the `llm` CLI with Perplexity plugin for web research, deep reasoning, and independent validation.
4
5	## Quick Start
6
7	```bash
8	llm -m sonar 'your question here'
9	```
10
11	## Model Selection
12
13	\| Model \| Use Case \| Context \| Web Search \|
14	\|-------\|----------\|---------\|------------\|
15	\| `sonar` \| Quick lookups, general questions (default) \| 128k \| ✓ \|
16	\| `sonar-pro` \| Thorough answers, detailed research \| 200k \| ✓ \|
17	\| `sonar-deep-research` \| Complex multi-faceted research \| 128k \| ✓ \|
18	\| `sonar-reasoning` \| Step-by-step reasoning questions \| 128k \| ✓ \|
19	\| `sonar-reasoning-pro` \| Advanced reasoning with detail \| 128k \| ✓ \|
20	\| `r1-1776` \| Reasoning without web search \| 128k \| ✗ \|
21
22	Start with `sonar` for most queries. Upgrade to other models when you need comprehensive coverage (`sonar-pro`), complex synthesis (`sonar-deep-research`), or logical reasoning (`sonar-reasoning`/`sonar-reasoning-pro`).
23
24	For detailed selection criteria, decision flowcharts, and examples, see [references/model-selection-guide.md](references/model-selection-guide.md).
25
26	## When to Use
27
28	Use this skill in three primary scenarios:
29
30	### 1. Current or External Knowledge
31
32	When you need up-to-date information and intentionally want to use Perplexity instead of standard web fetch tools:
33
34	- Documentation lookups: Current API docs, library references, framework guides
35	- Library versions: "What's the latest version of X?"
36	- Recent changes: "What changed in Y version 2.0?"
37	- Error research: Unfamiliar error messages or stack traces
38	- Technology verification: "Does X support Y feature?"
39	- Current events: Recent news, announcements, releases
40	- Fact checking: Information that may have changed since training cutoff
41
42	### 2. Second Opinions
43
44	When you want an independent perspective before committing to a decision or approach:
45
46	- Architecture decisions: "What are the trade-offs between approach A and B?"
47	- Technology choices: "Should I use X or Y for this use case?"
48	- Design validation: "Is this approach considered best practice?"
49	- Problem verification: "Am I thinking about this problem correctly?"
50	- Alternative perspectives: "What other ways could this be solved?"
51
52	### 3. Deep Reasoning Delegation
53
54	When you want to offload complex analytical thinking to specialized reasoning models (even when information doesn't need to be current):
55
56	- Complex problem solving: Multi-step reasoning requiring systematic analysis
57	- Algorithmic thinking: "How should I optimize this algorithm?"
58	- System design: "How should I architect this system?"
59	- Trade-off analysis: Comparing multiple options with many variables
60	- Mathematical reasoning: Calculations, proofs, or quantitative analysis
61
62	## When NOT to Use
63
64	Do NOT use for:
65
66	- Local codebase questions: Use grep/read tools instead
67	- Information already in context: Use existing conversation context
68	- Simple factual questions: Use your training data when sufficient
69	- Tasks requiring file operations: Code analysis, refactoring, editing
70
71	This skill complements, not replaces, standard web tools (WebFetch, WebSearch). Choose based on whether you want Perplexity's specific capabilities (reasoning models, synthesis, alternative perspective).
72
73	## Query Best Practices
74
75	Craft effective queries for better results:
76	- Be specific and detailed rather than vague
77	- Include time constraints for current information
78	- Specify output format when helpful
79	- Use domain-appropriate terminology
80
81	For comprehensive query strategies and examples, see [references/query-strategies.md](references/query-strategies.md).
82
83	## Usage Examples
84
85	### Current Knowledge: Documentation Lookup
86	```bash
87	llm -m sonar 'Next.js App Router documentation 2026'
88	```
89
90	### Current Knowledge: Error Research
91	```bash
92	llm -m sonar 'ECONNREFUSED error when connecting to PostgreSQL'
93	```
94
95	### Current Knowledge: Library Version
96	```bash
97	llm -m sonar 'latest stable version of React 2026'
98	```
99
100	### Current Knowledge: Complex Research
101	```bash
102	llm -m sonar-deep-research 'comparison of WebSocket vs Server-Sent Events for real-time updates in 2026'
103	```
104
105	### Second Opinion: Architecture Decision
106	```bash
107	llm -m sonar-pro 'trade-offs between microservices and monolith for a team of 5 engineers'
108	```
109
110	### Second Opinion: Technology Choice
111	```bash
112	llm -m sonar 'should I use PostgreSQL or MongoDB for a social media analytics platform'
113	`

People who liked this also viewed

6 picks

Type

perplexity-web-research

96installs2forks

View on GitHub

How to install perplexity-web-research?

xpepper/perplexity-agent-skill/perplexity-web-research

$npx -y skills add xpepper/perplexity-agent-skill --skill perplexity-web-research

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Perplexity Web Research
2
3	Use the `llm` CLI with Perplexity plugin for web research, deep reasoning, and independent validation.
4
5	## Quick Start
6
7	```bash
8	llm -m sonar 'your question here'
9	```
10
11	## Model Selection
12
13	\| Model \| Use Case \| Context \| Web Search \|
14	\|-------\|----------\|---------\|------------\|
15	\| `sonar` \| Quick lookups, general questions (default) \| 128k \| ✓ \|
16	\| `sonar-pro` \| Thorough answers, detailed research \| 200k \| ✓ \|
17	\| `sonar-deep-research` \| Complex multi-faceted research \| 128k \| ✓ \|
18	\| `sonar-reasoning` \| Step-by-step reasoning questions \| 128k \| ✓ \|
19	\| `sonar-reasoning-pro` \| Advanced reasoning with detail \| 128k \| ✓ \|
20	\| `r1-1776` \| Reasoning without web search \| 128k \| ✗ \|
21
22	Start with `sonar` for most queries. Upgrade to other models when you need comprehensive coverage (`sonar-pro`), complex synthesis (`sonar-deep-research`), or logical reasoning (`sonar-reasoning`/`sonar-reasoning-pro`).
23
24	For detailed selection criteria, decision flowcharts, and examples, see [references/model-selection-guide.md](references/model-selection-guide.md).
25
26	## When to Use
27
28	Use this skill in three primary scenarios:
29
30	### 1. Current or External Knowledge
31
32	When you need up-to-date information and intentionally want to use Perplexity instead of standard web fetch tools:
33
34	- Documentation lookups: Current API docs, library references, framework guides
35	- Library versions: "What's the latest version of X?"
36	- Recent changes: "What changed in Y version 2.0?"
37	- Error research: Unfamiliar error messages or stack traces
38	- Technology verification: "Does X support Y feature?"
39	- Current events: Recent news, announcements, releases
40	- Fact checking: Information that may have changed since training cutoff
41
42	### 2. Second Opinions
43
44	When you want an independent perspective before committing to a decision or approach:
45
46	- Architecture decisions: "What are the trade-offs between approach A and B?"
47	- Technology choices: "Should I use X or Y for this use case?"
48	- Design validation: "Is this approach considered best practice?"
49	- Problem verification: "Am I thinking about this problem correctly?"
50	- Alternative perspectives: "What other ways could this be solved?"
51
52	### 3. Deep Reasoning Delegation
53
54	When you want to offload complex analytical thinking to specialized reasoning models (even when information doesn't need to be current):
55
56	- Complex problem solving: Multi-step reasoning requiring systematic analysis
57	- Algorithmic thinking: "How should I optimize this algorithm?"
58	- System design: "How should I architect this system?"
59	- Trade-off analysis: Comparing multiple options with many variables
60	- Mathematical reasoning: Calculations, proofs, or quantitative analysis
61
62	## When NOT to Use
63
64	Do NOT use for:
65
66	- Local codebase questions: Use grep/read tools instead
67	- Information already in context: Use existing conversation context
68	- Simple factual questions: Use your training data when sufficient
69	- Tasks requiring file operations: Code analysis, refactoring, editing
70
71	This skill complements, not replaces, standard web tools (WebFetch, WebSearch). Choose based on whether you want Perplexity's specific capabilities (reasoning models, synthesis, alternative perspective).
72
73	## Query Best Practices
74
75	Craft effective queries for better results:
76	- Be specific and detailed rather than vague
77	- Include time constraints for current information
78	- Specify output format when helpful
79	- Use domain-appropriate terminology
80
81	For comprehensive query strategies and examples, see [references/query-strategies.md](references/query-strategies.md).
82
83	## Usage Examples
84
85	### Current Knowledge: Documentation Lookup
86	```bash
87	llm -m sonar 'Next.js App Router documentation 2026'
88	```
89
90	### Current Knowledge: Error Research
91	```bash
92	llm -m sonar 'ECONNREFUSED error when connecting to PostgreSQL'
93	```
94
95	### Current Knowledge: Library Version
96	```bash
97	llm -m sonar 'latest stable version of React 2026'
98	```
99
100	### Current Knowledge: Complex Research
101	```bash
102	llm -m sonar-deep-research 'comparison of WebSocket vs Server-Sent Events for real-time updates in 2026'
103	```
104
105	### Second Opinion: Architecture Decision
106	```bash
107	llm -m sonar-pro 'trade-offs between microservices and monolith for a team of 5 engineers'
108	```
109
110	### Second Opinion: Technology Choice
111	```bash
112	llm -m sonar 'should I use PostgreSQL or MongoDB for a social media analytics platform'
113	`

People who liked this also viewed

6 picks

Type