fusion-research

Source-backed research orchestrator for the Fusion ecosystem. Routes to the correct research agent based on question type. Returns source-backed evidence only; will not invent Framework behavior, component APIs, skill catalog relationships, platform guidance, or backend implementation details. USE FOR: any research question needing source-backed evidence about Fusion Framework APIs, EDS components, the Fusion skill catalog, Fusion platform documentation, or backend service implementation. DO NOT USE FOR: implementing code changes, installing or editing skills, MCP setup or troubleshooting, or inventing Fusion behavior without evidence.

Quality

63%

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/fusion-research/SKILL.md

Quality

Content

38%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 is well-organized and clearly sequenced, but its central dispatch mechanism references an agents/ directory that is absent from the bundle, while the supplied references/*.query.md files are never linked, leaving the most important navigation paths broken or undiscoverable.

Suggestions

Add the missing agents/*.agent.md files (or repoint the Step 1/Step 2/'Research agents' references to the existing references/*.query.md files) so the dispatch step is actually executable.

Link each references/*.query.md file from the corresponding agent or domain entry so the supplied query patterns are discoverable rather than orphaned.

De-duplicate the agent listing (it appears in the Step 1 table, Step 2 list, and 'Research agents' section) and trim the 14 near-identical trigger bullets to a tighter, distinct set.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of basic-concept padding, but the five research agents are listed three times (Step 1 table, Step 2 dispatch list, and the 'Research agents' section) and the 14 'Typical triggers' bullets repeat near-identical phrasings, so it could be tightened.	3 / 5
Actionability	The core dispatch instruction ('follow agents/framework.agent.md' etc.) points at five named agent files that do not exist in the bundle, so the specific steps cannot be executed as written; the existing references/*.query.md files that would carry the query patterns are never referenced from the body.	2 / 5
Workflow Clarity	Steps 0-3 are clearly sequenced (sharpen scope, classify, dispatch, return), but the validation checkpoints ('capture the source path and supporting excerpt before finalizing', 'state uncertainty after one refinement pass') live in a separate 'Safety & constraints' section rather than being woven into the steps, so checkpoints are implicit rather than inline.	3 / 5
Progressive Disclosure	Asset references (source-backed-answer-template.md, four follow-up files) are valid and one-level-deep, but the five central agents/.agent.md references point to a non-existent directory (broken navigation for the core workflow) and the five references/.query.md files are orphaned — present in the bundle but never linked from the body.	2 / 5
	Total	10 / 20 Passed

Description

87%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 strong, well-structured description that clearly states both capability and trigger conditions with explicit use/avoid boundaries and third-person voice. Minor room to add synonym-style trigger variants and broaden the concrete action list beyond route-and-return.

Dimension	Reasoning	Score
Specificity	Names the domain ('Fusion ecosystem') and concrete actions ('Routes to the correct research agent based on question type', 'Returns source-backed evidence only') across five sub-domains, but the action verbs are limited to routing and evidence-return rather than the many distinct operations in the score-5 anchor.	4 / 5
Completeness	Explicitly answers both 'what' ('Source-backed research orchestrator... Routes to the correct research agent... Returns source-backed evidence only') and 'when' ('USE FOR: any research question needing source-backed evidence about...'), with concrete trigger phrases and negative boundaries; uses third person ('Routes', 'Returns') so no voice penalty applies.	5 / 5
Trigger Term Quality	Strong keyword coverage with terms users in this ecosystem would naturally say ('research question', 'Fusion Framework APIs', 'EDS components', 'skill catalog', 'platform documentation', 'backend service implementation'), but no synonym or file-extension variants as required for a 5.	4 / 5
Distinctiveness Conflict Risk	Clear niche (Fusion-ecosystem source-backed research) with distinct triggers, and an explicit 'DO NOT USE FOR' clause separating it from dev, skill-discovery, authoring, and MCP skills, minimizing conflict risk.	5 / 5
	Total	18 / 20 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
metadata_field	'metadata' should map string keys to string values	Warning
relative_links	Relative link issues: 15 missing	Warning

	Total	14 / 16 Passed

Repository: equinor/fusion-framework
Path: .agents/skills/fusion-research/SKILL.md
Commit: bf61e57

Reviewed: 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.