write

Use when a quest has enough evidence to draft or refine a paper, report, or research summary without inventing missing support.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./src/skills/write/SKILL.md

Quality

Content

77%

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 strong, validated workflows and well-signaled one-level-deep references, but it is over-long with duplicated guidance, inlines a major sub-skill that should be its own reference, and contains two broken reference paths. De-duplicating and externalizing the oral-upgrade material would materially improve it.

Suggestions

Move the 'Draft To Top Conference Oral' section into its own reference file and keep only a short routing pointer in the main body to cut the ~813-line monolith down to an overview.

De-duplicate the repeated cleanliness/provenance rules that recur across 'AVOID/Pitfalls', 'Constraints', 'Keep Manuscript Text Clean', and 'Common Failure Modes' into a single authoritative list.

Fix or remove the dangling references on line 151 (`references/outline-evidence-contract-example.md` and `references/paper-experiment-matrix-template.md`), which do not exist in the bundle.

Dimension	Reasoning	Score
Conciseness	The prose is dense and free of basic-concept filler, but at ~813 lines it is verbose: guidance is repeated across 'AVOID/Pitfalls', 'Constraints', 'Keep Manuscript Text Clean', and 'Common Failure Modes', and the large 'Draft To Top Conference Oral' section duplicates themes from the main body. It could be tightened substantially without losing clarity.	2 / 3
Actionability	Provides copy-paste-ready tool calls with parameters (e.g. `artifact.get_paper_contract(detail='full')`, `artifact.submit_paper_outline(mode='candidate'\|'select'\|'revise', ...)`), concrete file paths, and explicit section-generation orders, giving fully executable guidance.	3 / 3
Workflow Clarity	The 9-step Workflow is clearly sequenced with explicit validation checkpoints (`validate_academic_outline`, `validate_manuscript_language`, `validate_manuscript_coverage`) and feedback loops (validate -> repair -> re-validate, publishability stop-loss routing to `decision`) appropriate for batch/destructive submission operations.	3 / 3
Progressive Disclosure	Four referenced files (oral_package_patterns, oral_writing_principles, experiments_analysis_patterns, section_rewrite_checklist) exist one level deep and are clearly signaled, but the body is a monolithic 813-line wall that inlines the large oral-upgrade section instead of splitting it out, and line 151 cites two references (outline-evidence-contract-example.md, paper-experiment-matrix-template.md) that do not exist.	2 / 3
	Total	10 / 12 Passed

Description

75%

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 well-structured with an explicit trigger and a distinctive evidence-gating condition, but its action vocabulary is sparse and slightly jargon-laden. Tightening specificity and adding more natural user phrasings would lift it toward the top anchors.

Suggestions

Expand the action list beyond 'draft or refine' to include concrete verbs like 'prepare a paper bundle', 'revise a section', or 'upgrade a draft toward submission'.

Replace internal jargon ('quest', 'enough evidence') or pair it with natural user terms ('draft a paper', 'revise my draft', 'write a research summary').

Add common trigger variations such as 'write up', 'tighten a report', or 'package a submission' to broaden keyword coverage.

Dimension	Reasoning	Score
Specificity	Names the domain and a few concrete actions ('draft or refine a paper, report, or research summary') plus an evidence constraint, but the action list is terse rather than a comprehensive set of multiple specific actions, matching the 'names domain and some actions' anchor.	2 / 3
Completeness	An explicit 'Use when...' clause gives the trigger condition while the same sentence states what the skill does (draft/refine papers, reports, summaries), answering both 'when' and 'what'.	3 / 3
Trigger Term Quality	Includes natural phrases a user might say ('draft or refine a paper', 'research summary') but leans on internal jargon ('quest', 'enough evidence') and omits common variations like 'write up', 'revise a draft', or 'prepare a submission'.	2 / 3
Distinctiveness Conflict Risk	The evidence-gate condition ('without inventing missing support') scopes it to evidence-backed academic drafting and separates it from generic writing or literature-search skills, giving it a clear niche.	3 / 3
	Total	10 / 12 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
skill_md_line_count	SKILL.md is long (814 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: ResearAI/DeepScientist
Path: src/skills/write/SKILL.md
Commit: b366244

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.