bgpt-mcp

Search scientific papers via the BGPT MCP server and retrieve structured experimental data — methods, results, conclusions, quality scores, and 25+ metadata fields per paper.

Quality

71%

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/bgpt-mcp/SKILL.md

Quality

Content

71%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 highly actionable with executable MCP config, tool params, and CLI examples, and the workflow is clearly sequenced. Its weaknesses are conciseness (padded background sections) and progressive disclosure (reference material inlined rather than split into one-level-deep files).

Suggestions

Move the CLI flag table, full JSON output schema, and pricing table into a single reference file (e.g. REFERENCE.md) and link to it from the body to improve progressive disclosure.

Trim the 'Why This Exists' and 'Algorithm/Methodology' sections to one or two lines each, or fold the key points into Gotchas, since they restate context Claude can infer.

Inline the 'call failed or returned no results → say so; do not invent data' checkpoint as an explicit step in the Workflow rather than only in Gotchas.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with concrete config and examples, but padded sections ('Why This Exists' prose, a restated trigger list that duplicates frontmatter, and an 'Algorithm/Methodology' ingestion pipeline that is background rather than action) could be trimmed without losing clarity.	3 / 5
Actionability	Provides copy-paste-ready MCP client JSON, the exact tool name and params, CLI commands with a flag table, and a full example output block — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	The 5-step Workflow is clearly sequenced (parse → connect → retrieve → present → attribute) and the skill is non-destructive (search/retrieval), so the destructive-cap does not apply; however, the 'call failed → say so' validation lives in Gotchas rather than as an inline checkpoint, leaving a minor validation gap.	4 / 5
Progressive Disclosure	No bundle files exist and all reference material (MCP config, CLI flag table, JSON output schema, pricing table) is inlined in one ~250-line file; headers give some structure, but content that clearly belongs in separate reference files is not split out.	3 / 5
	Total	15 / 20 Passed

Description

71%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.

The description is specific and distinctive, naming concrete actions and the structured data fields returned, but it lacks an explicit 'when to use' trigger clause, which caps its completeness. Adding a 'Use when the user asks to search papers / find experimental data...' clause would raise the completeness and overall score.

Suggestions

Append an explicit trigger clause, e.g. 'Use when the user asks to search papers, find studies, or retrieve experimental data (methods, results, quality scores) from full-text scientific literature.'

Add a distinguishing cue against PubMed-abstract skills, e.g. 'prefer this over PubMed-summariser when full-text methods/results/quality are needed.'

Include one or two natural synonyms already in trigger_keywords (e.g. 'find papers', 'research papers') directly in the description text.

Dimension	Reasoning	Score
Specificity	Enumerates concrete actions ('Search scientific papers', 'retrieve structured experimental data') and lists specific deliverables — 'methods, results, conclusions, quality scores, and 25+ metadata fields per paper' — giving comprehensive coverage of capabilities.	5 / 5
Completeness	It clearly states 'what' (search + retrieve structured experimental data with listed fields) but provides no explicit 'Use when...' trigger clause; per the rubric, a missing explicit trigger guidance caps completeness at 3.	3 / 5
Trigger Term Quality	The description leads with the natural phrase 'Search scientific papers' and names the domain ('BGPT MCP server', 'experimental data'), but stops short of the synonym/file-extension breadth (e.g. PubMed, .pdf-equivalent handles) seen in the top anchor.	4 / 5
Distinctiveness Conflict Risk	The full-text-experimental-data niche via the named 'BGPT MCP server' is clearly distinct from PubMed-abstract skills, leaving only minor overlap risk with closely related literature skills.	4 / 5
	Total	16 / 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
metadata_field	'metadata' should map string keys to string values	Warning

	Total	15 / 16 Passed

Repository: ClawBio/ClawBio
Path: skills/bgpt-mcp/SKILL.md
Commit: d30153a

Reviewed: 3 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.