perplexity-research

Brain-augmented web research. Sends brain context about a topic to Perplexity, which searches the web with citations and returns what is NEW vs what the brain already knows. Use for entity enrichment, current-state checks, deal monitoring, and freshness deltas. NOT for simple URL fetches (use web_fetch) or brain-only queries (use gbrain query).

Quality

75%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/perplexity-research/SKILL.md

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An actionable, well-structured skill body with concrete commands and a clear invocation sequence. It loses points for test-conformance boilerplate padding, missing validation checkpoints in the workflow, and a monolithic single-file structure with no progressive split.

Suggestions

Add an explicit validation checkpoint in the Invocation workflow, e.g. step 4.5: verify every claim in the Perplexity response has a citation URL before calling put_page.

Remove or merge the Contract and Output Format conformance-test stubs so every section earns its tokens, or fold their guarantees into existing sections.

Consider splitting the full output template and model table into a references/ file (e.g. OUTPUT_FORMAT.md) referenced from a concise Quick start, to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and assumes Claude's competence (no explaining what Perplexity or citations are), but the trailing Contract and Output Format sections are explicit test-conformance stubs ("this section exists for the conformance test") and the output frontmatter is restated, so not every token earns its place. Not a 3 because of that non-earning boilerplate; not a 1 because the core content is efficient.	2 / 3
Actionability	Provides an executable curl call with auth header and JSON body, real gbrain commands, a concrete model+cost table, and specific recency_filter values. Not a 2 because the guidance is copy-paste ready rather than pseudocode.	3 / 3
Workflow Clarity	The Invocation section gives a clear 5-step numbered sequence (pull context, compose query, call API, write page, cross-link), but there is no explicit validation checkpoint (e.g., verify every claim has a citation before put_page) despite citation completeness being a stated requirement. Not a 3 because checkpoints are implicit; not a 1 because the sequence is clearly laid out.	2 / 3
Progressive Disclosure	Well-organized single file with well-signaled one-level-deep cross-skill/convention references and no nested references, but no bundle files exist and all content is inline in one ~180-line document with no quick-start-vs-advanced split. Not a 3 because content is not progressively split; not a 1 because references are shallow and clearly signaled.	2 / 3
	Total	9 / 12 Passed

Description

85%

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, specific description that clearly states what the skill does, when to use it, and what to use instead for adjacent needs. Its main weakness is that the most natural user trigger phrases live in the triggers list rather than the description text, slightly limiting trigger-term quality.

Dimension	Reasoning	Score
Specificity	"Sends brain context about a topic to Perplexity, which searches the web with citations and returns what is NEW vs what the brain already knows" lists multiple concrete actions (send context, search with citations, return the delta). Not a 2 because it goes beyond naming a domain to enumerate specific operations.	3 / 3
Completeness	Explicitly answers what ("returns what is NEW vs what the brain already knows") and when ("Use for entity enrichment, current-state checks, deal monitoring, and freshness deltas"), plus exclusions ("NOT for simple URL fetches... or brain-only queries"). Clearly covers both what and when.	3 / 3
Trigger Term Quality	Includes relevant terms like "web research", "entity enrichment", "current-state checks", "deal monitoring", and "freshness deltas", but the most natural user phrasings ("what's new about", "what changed about") appear only in the triggers list, not the description. Not a 3 because common conversational variations are missing from the description text itself.	2 / 3
Distinctiveness Conflict Risk	Clear niche (brain-augmented web research via Perplexity) with explicit boundary-setting against web_fetch and gbrain query, making it unlikely to trigger for the wrong skill. Not a 2 because the exclusions remove overlap with adjacent tools.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing, 2 suspicious	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/perplexity-research/SKILL.md
Commit: 3fafb69

Reviewed: about 20 hours 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.