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

77%

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-structured, highly actionable workflow: concrete MCP tool calls with parameters, quantified search strategy, a complete report template, and a quality-rules checklist. The main improvement areas are making the parallel-subagent section executable (actual Task tool syntax) and moving the long report template into a reference file.

Suggestions

Replace the pseudocode block in 'Parallel Research with Subagents' with an example of the actual Task/Agent tool invocation including the prompt each research agent should receive.

Move the full markdown report template into a references/report-template.md file and link to it, keeping only the section list inline in SKILL.md.

Merge the 'Examples' section into 'When to Activate' (they duplicate each other) to trim tokens, and add an explicit Step 6 checkpoint such as verifying every claim has a citation before delivering.

Dimension	Reasoning	Score
Conciseness	The body is lean and imperative and assumes Claude's competence throughout, with only minor trims available: the 'Examples' section largely duplicates the 'When to Activate' trigger list, and 'Both together give the best coverage' adds little — efficient with minor instances of over-explanation rather than the every-token-earns-its-place 5.	4 / 5
Actionability	Concrete, copy-paste-ready MCP call signatures with real parameters (firecrawl_search limit: 8, crawling_exa tokensNum: 5000) plus numeric targets (2-3 keyword variations, 15-30 sources, 3-5 deep reads) make this mostly executable; the parallel-research section never shows the actual Task tool invocation and 'Configure in ~/.claude.json' gives no how, which are the minor gaps that keep it below fully-executable.	4 / 5
Workflow Clarity	Six clearly sequenced steps with a skip condition in Step 1 and delivery branching in Step 6 form a clear sequence; research is not a destructive or batch operation so no cap applies, but there is no explicit verify-report-before-delivering checkpoint, matching 'clear sequence with most checkpoints present; minor validation gaps' rather than the explicit-validation 5.	4 / 5
Progressive Disclosure	There are no bundle files (references/, scripts/, assets/ are absent) and the ~150-line body is well-sectioned with clear headers; the ~30-line report template is arguably content that belongs in a reference file, which is a minor organization gap placing it at good-structure rather than the appropriately-split 5.	4 / 5
	Total	16 / 20 Passed

Description

80%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 description with concrete, third-person capability statements, named tools, and an explicit 'Use when' clause. Its main weakness is trigger coverage: the natural phrases users actually say for research requests (deep dive, investigate, current state of) appear in the body but not the description itself.

Suggestions

Add the trigger phrases already listed in the body's 'When to Activate' section to the description, e.g. 'Use when the user says research, deep dive, investigate, or what's the current state of a topic.'

Sharpen the 'when' clause from 'on any topic with evidence and citations' toward concrete scenarios like competitive analysis, technology evaluation, or market sizing to reduce overlap with generic web-search skills.

Dimension	Reasoning	Score
Specificity	"Searches the web, synthesizes findings, and delivers cited reports with source attribution" lists multiple concrete actions, and naming the firecrawl/exa MCPs makes coverage comprehensive within the domain — no meaningful gap that would justify a 4.	5 / 5
Completeness	Both a clear "what" (search, synthesize, deliver cited reports) and an explicit "Use when the user wants thorough research on any topic with evidence and citations" are present; the "when" is explicit but generic rather than listing concrete trigger phrases, matching the anchor where 'when' could be more specific rather than the 5 anchor.	4 / 5
Trigger Term Quality	"thorough research", "citations", and "evidence" are natural user terms, but common phrasings like "deep dive", "investigate", or "what's the current state of" — which the body itself lists as triggers — are missing from the description, so it falls just short of the comprehensive-synonyms anchor.	4 / 5
Distinctiveness Conflict Risk	Naming firecrawl and exa MCPs carves out a distinct tooling niche, but "research on any topic" is broad and would overlap with any general research or web-search skill — minor overlap risk with closely related skills, not the minimal-conflict 5 anchor.	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: c70874f

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.