best-practices-researcher

Researches and synthesizes external best practices, documentation, and examples for any technology or framework. Use when you need industry standards, community conventions, or implementation guidance.

Quality

55%

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 ./.agents/skills/best-practices-researcher/SKILL.md

Quality

Content

48%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.

The body's core structure is strong: an explicitly ordered three-phase methodology with a mandatory deprecation-check gate and a source-attribution hierarchy that are genuinely non-obvious. It is dragged down by significant padding — a role preamble and generic research advice that assume Claude lacks basic competence — and by abstract delivery guidance where an example output would do more work.

Suggestions

Cut the role preamble and generic research bullets ("Search the web for recent articles…", "Consider the recency of information…") — keep only the non-obvious: the phase order, the deprecation gate, the skill mappings, and the attribution hierarchy.

Replace the abstract "Deliver Actionable Guidance" bullets with one short example of the expected output format (categorized findings with a labeled source line), which is more actionable than describing the format.

Move the skill-directory path listing, the technology-to-skill mapping table, and the GitHub-issues special case into a reference file (e.g. references/skill-mappings.md), keeping SKILL.md as a lean overview with clearly signaled links.

Dimension	Reasoning	Score
Conciseness	Several padded sections restate what a competent model does by default: the role preamble ("You are an expert technology researcher specializing in discovering, analyzing, and synthesizing…"), generic bullets like "Search the web for recent articles, guides, and community discussions" and "Organize into clear categories", and instructions on how to use glob/file-read. Not a level-3 "some unnecessary explanation" case — the padding is distributed across most sections.	2 / 5
Actionability	Genuinely concrete elements exist — exact search query templates ("[API name] deprecated [current year] sunset shutdown"), specific skill-directory globs, and a technology-to-skill mapping table — but much of the guidance stays abstract ("Extract best practices, code patterns, and conventions", "Present findings in a structured, easy-to-implement format") with no examples of the output format being prescribed. Falls between the incomplete-guidance anchor and the mostly-executable anchor, closer to 3.	3 / 5
Workflow Clarity	The four phases are explicitly ordered ("Research Methodology (Follow This Order)"), Phase 1 ends with a coverage decision point, and Phase 1.5 is a hard validation gate ("Report findings before proceeding - do not recommend deprecated APIs"). Not a 5 because the synthesis and delivery phases have no checkpoints or feedback loops — nothing says how to verify the research answered the user's question or what to do when sources conflict beyond "present the different viewpoints".	4 / 5
Progressive Disclosure	No bundle files exist, and the skill is a single ~125-line file. Section headers give it structure, but content that clearly belongs in a reference file is inlined: the six-path skill-directory listing, the technology-to-skill mapping table, and the GitHub-issues special case could each be a one-level-deep reference, which would also let the body serve as a leaner overview.	3 / 5
	Total	12 / 20 Passed

Description

62%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 grammatically clean, third-person description that correctly pairs a what-statement with an explicit when-statement. Its weaknesses are generic research verbs in place of concrete capabilities and an unbounded "any technology or framework" scope that gives it high overlap risk with other research and documentation skills.

Suggestions

Replace the generic verbs with the concrete activities the agent actually performs, e.g. "Discovers local skills, verifies API deprecation status, and synthesizes findings with source attribution."

Narrow the trigger from "any technology or framework" to the situations where this agent is the right choice (e.g. "Use when the user asks for best practices, conventions, or standards for a specific technology before implementing it").

Add natural user phrasings to the when-clause, such as "what's the best way to…", "current best practices for…", or "recommended conventions for…".

Dimension	Reasoning	Score
Specificity	Names the domain ("external best practices, documentation, and examples") and two actions ("Researches and synthesizes"), but the actions are generic research verbs rather than the multi-action coverage of the level-4/5 anchors.	3 / 5
Completeness	Both parts are present: a clear what ("Researches and synthesizes external best practices, documentation, and examples") and an explicit when ("Use when you need industry standards, community conventions, or implementation guidance"). The when-clause is explicit but abstract — it lacks the concrete trigger phrases (specific user utterances, file types, task shapes) of a level-5 description.	4 / 5
Trigger Term Quality	Includes natural phrases users would say — "industry standards", "community conventions", "implementation guidance", "best practices" — but misses common synonyms and phrasings like "style guides", "how should I structure", or specific technology mentions.	4 / 5
Distinctiveness Conflict Risk	"for any technology or framework" makes the scope unbounded, and "researches and synthesizes" documentation is the core function of most research/documentation skills, so it would fire on nearly any research-flavored request rather than a distinct niche.	2 / 5
	Total	13 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/best-practices-researcher/SKILL.md
Commit: babb3c2

Reviewed: 2 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.