deep-research

Multi-source deep research using firecrawl and exa MCPs. Searches the web, synthesizes findings, and delivers cited reports with source attribution. Use when the user wants thorough research on any topic with evidence and citations.

Quality

79%

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/deep-research/SKILL.md

Quality

Content

75%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.

A well-organized, actionable research workflow skill with concrete MCP tool-call examples and a ready-to-use report template. The main gaps are minor: a few verbose prose passages, and validation guidance that lives in a separate Quality Rules section rather than being wired in as an explicit workflow checkpoint.

Suggestions

Tighten or remove the scripted clarifying-questions dialogue in Step 1 and trim filler prose like 'Both together give the best coverage' to improve token efficiency.

Fold a brief verification checkpoint into the workflow — e.g. before Step 6 Deliver, add 'Verify: every claim has a source; flag single-source claims as unverified' — to raise workflow clarity.

Consider moving the full report markdown template or the Examples list into a referenced bundle file to deepen progressive disclosure, since the body exceeds 50 lines.

Dimension	Reasoning	Score
Conciseness	The body is generally lean with sectioned, purposeful content (workflow steps, report template, quality rules), but contains minor over-explanation such as the scripted clarifying-questions dialogue and prose like 'Both together give the best coverage' that could be trimmed.	4 / 5
Actionability	Provides concrete, executable MCP call examples with real parameters (e.g. web_search_advanced_exa with startPublishedDate, crawling_exa with tokensNum: 5000) and a copy-paste report template, but interleaves these with higher-level strategy guidance ('Aim for 15-30 unique sources', 'Use 2-3 keyword variations') that keeps it just shy of fully executable.	4 / 5
Workflow Clarity	Six clearly sequenced steps with a separate Quality Rules section that functions as a verification checklist (cross-reference, flag unverified, acknowledge gaps), but validation is not embedded as an explicit checkpoint within the workflow sequence itself, leaving a minor gap relative to the 5 anchor.	4 / 5
Progressive Disclosure	Well-organized into clearly labeled sections (When to Activate, MCP Requirements, Workflow, Parallel Research, Quality Rules, Examples) in a single flat file with no nested references; at ~150 lines with the full report template and examples inlined it is well-structured but not split, sitting at 4 rather than 5.	4 / 5
	Total	16 / 20 Passed

Description

83%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 concrete actions and named tooling. Its main weakness is trigger-term breadth — it relies on 'research' and lacks the synonyms ('deep dive', 'investigate') that would make activation more reliable.

Suggestions

Add synonyms to the trigger clause, e.g. 'Use when the user wants thorough research, a deep dive, or an investigation on any topic with evidence and citations.'

Consider noting the deliverable format more distinctly (e.g. 'delivers a structured cited report') to sharpen the 'what' beyond the current action list.

Dimension	Reasoning	Score
Specificity	Names the domain and tooling ('using firecrawl and exa MCPs') and lists several concrete actions — 'Searches the web, synthesizes findings, and delivers cited reports with source attribution' — but stops short of the comprehensive multi-action coverage of a 5.	4 / 5
Completeness	Explicitly answers both 'what' (multi-source deep research, searches/synthesizes/delivers cited reports) and 'when' ('Use when the user wants thorough research on any topic with evidence and citations') with concrete trigger phrasing, matching the 5 anchor.	5 / 5
Trigger Term Quality	Includes the natural trigger 'thorough research on any topic with evidence and citations', but omits common synonyms a user would actually say (e.g. 'deep dive', 'investigate', 'look into') that appear only in the body, leaving a few natural terms missing.	4 / 5
Distinctiveness Conflict Risk	The firecrawl/exa MCP framing plus 'cited reports with source attribution' carves a fairly distinct niche, but 'research on any topic' is broad enough to risk minor overlap with general web-search or Q&A skills, so it sits at 4 rather than 5.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Path: .agents/skills/deep-research/SKILL.md
Commit: d8409a4

Reviewed: 1 day 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.