CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

academic-research-suite

ARS-Codex research, writing, manuscript review, study screening, and experiments. Use for deep research, literature or systematic reviews, meta-analysis, research questions, drafts, revisions, roadmaps, abstracts, citations, integrity checks, peer review, research-to-paper. Screen records: sr-screener. Citation triggers: check citations, look over the refs, 檢查引用, 檢查參考文獻, 인용 확인, 인용 형식 검사. Korean: 논문 심사, 논문 수정, 초록 작성, 체계적 문헌고찰, 연구부터 논문까지. Español: revisión de literatura, revisar artículo, enmendar mi artículo, escribir resumen, investigación a artículo. Also use for ARS aliases: /ars-plan, /ars-outline, /ars-abstract, /ars-lit-review, /ars-citation-check, /ars-disclosure, /ars-format-convert, /ars-3w, /ars-revision-coach, /ars-revision, /ars-reviewer, /ars-mark-read, /ars-unmark-read, /ars-cache-invalidate, /ars-rebuttal-audit, /ars-full. Prompts, references, templates, and contracts live under ars/.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Failed to scan

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./plugins/ars-codex/skills/academic-research-suite/SKILL.md

The canonical home for this skill is academic-research-suite in Imbad0202/academic-research-skills-codex

SKILL.md
Quality
Evals
Security

Quality

Content

56%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 a highly actionable, well-gated router with exact paths, commands, and fail-closed validation rules, but it is far too heavy for a root SKILL.md: version-specific release contracts and duplicated referenced content are inlined rather than deferred to the bundle files it already points to. This hurts token efficiency and works against its own excellent progressive-disclosure First Rule.

Suggestions

Move the version-specific sections (v3.24.x, v3.23.x, v3.22.x release behavior) into a changelog or reference file (e.g. ars/docs/ or codex/), keeping only the current behavioral rules in the router to improve conciseness.

Stop inlining the routing core duplicated from ars/shared/references/routing_core.md — reference it and keep only a short summary, so the router is one level of indirection rather than a copy.

Trim the environment-variable and cross-model transport mapping rows to the commonly needed subset and defer the full contracts to ars/shared/cross_model_verification.md and codex/model-runtime-policy.md, which the body already cites.

DimensionReasoningScore

Conciseness

The ~650-line (52KB) root router inlines release-note sections ("ARS v3.24.2 Experiment Locks", "ARS v3.24.1 Reviewer Card Confirmation", "ARS v3.24.0 Review-Letter Attachments", "ARS v3.22.0 Contract-Honesty Boundaries") dense with version numbers, plus an exhaustive environment-variable/transport mapping table — noticeably more than a router needs, fitting anchor 2's 'several unnecessary or padded sections' rather than anchor 3's 'some unnecessary explanation', though it never explains concepts Claude already knows (so not 1).

2 / 5

Actionability

Concrete, executable guidance throughout: routing tables with exact paths ("ars/deep-research/WORKFLOW.md"), exact filenames, and runnable commands ("ars/scripts/run_ledger.py append", "report --claims <file>", "ars/scripts/check_acronyms.py"); some directives remain abstract policy ("Preserve dissent with an evidence-based disposition") without concrete mechanics, keeping it below anchor 5.

4 / 5

Workflow Clarity

The routing core is a sequenced decision procedure (Steps 0-3) with clarification branches and an escape-hatch rule, and there are explicit validation gates and feedback loops ("Missing required inputs return HANDOFF_INCOMPLETE and block transition", fail-closed handoff parsing, "report the loading failure and stop that workflow"); minor gaps remain because the sheer density and heavy reliance on external WORKFLOW.md files obscure the top-level flow, so anchor 5 is not fully met.

4 / 5

Progressive Disclosure

The First Rule ("Do not load the whole suite by default… load only the agent, reference, template, or shared files needed") and clearly signaled one-level-deep references are excellent, but substantial content that belongs in separate files is inlined — the routing core block is admittedly "copied from ars/shared/references/routing_core.md" and "Canonical Agent Files" duplicates a directory listing — matching anchor 3's 'content that should be separate is inline' (above anchor 2, since references are not buried and structure is not minimal).

3 / 5

Total

13

/

20

Passed

Description

96%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 a strong, explicit what+when trigger surface: comprehensive capability enumeration, natural multilingual trigger phrases, and clear alias routing. Its only weakness is that several capability words are generic writing terms that slightly overlap with general writing/editing skills.

DimensionReasoningScore

Specificity

The description enumerates many concrete capabilities — "research, writing, manuscript review, study screening, and experiments" plus "deep research, literature or systematic reviews, meta-analysis, research questions, drafts, revisions, roadmaps, abstracts, citations, integrity checks, peer review, research-to-paper" — giving comprehensive coverage of the suite rather than 1-2 generic actions (anchor 4's 'minor gaps' understates it).

5 / 5

Completeness

It explicitly answers both 'what' (the enumerated research/writing/review/screening/experiment capabilities) and 'when' ("Use for deep research, literature or systematic reviews… Also use for ARS aliases: /ars-plan…") with concrete trigger phrases, matching the anchor-5 example structure exactly; a 4 would require the 'when' to be less explicit, which it is not.

5 / 5

Trigger Term Quality

Natural trigger phrases users would actually say appear with synonyms and in three languages: "check citations, look over the refs", "檢查引用, 檢查參考文獻", "인용 확인", "revisar artículo, enmendar mi artículo, escribir resumen", plus the /ars-* aliases; no common variant category is obviously missing (above anchor 4).

5 / 5

Distinctiveness Conflict Risk

The academic-research niche is clear and triggers like "manuscript review", "meta-analysis", "sr-screener" are distinct, but generic writing words ("drafts", "revisions", "roadmaps", "abstracts") create minor overlap risk with general writing/editing skills — 'mostly distinct; minor overlap risk' (anchor 4) rather than anchor 5's minimal conflict risk.

4 / 5

Total

19

/

20

Passed

Validation

68%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 11 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (669 lines); consider splitting into references/ and linking

Warning

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 1 missing

Warning

referenced_paths_exist

Referenced path issues: 7 missing

Warning

Total

11

/

16

Passed

Repository

Imbad0202/academic-research-skills-codex

Path

plugins/ars-codex/skills/academic-research-suite/SKILL.md

Reviewed

Table of Contents

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.