idea

Use when a quest needs concrete hypotheses, limitation analysis, candidate directions, or a selected idea relative to the active baseline.

Quality

56%

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 ./src/skills/idea/SKILL.md

Quality

Content

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill is highly actionable with concrete templates, durable paths, API calls, and well-checkpointed workflows, and it leverages real one-level-deep reference files. Its dominant weakness is severe verbosity: the same guidance is repeated across many overlapping sections, making the main file a monolithic wall of text rather than a lean overview.

Suggestions

Collapse the redundant workflow presentations ('Control workflow', 'Creative-divergence protocol', 'Integrated ideation workflow', 'Workflow') into a single canonical sequenced workflow; reference the others only for optional depth.

Move the repeated detailed rules (paper-floor counts, ideation lenses, failure-mode recovery, memory/artifact rules) into reference files and keep SKILL.md as a concise overview that points to them.

De-duplicate the 'why now / what changed', bounded-divergence-to-2-3-frontier, and selection-gate guidance that currently appears in 4-5 places.

Dimension	Reasoning	Score
Conciseness	The body is a ~1500-line wall of text that restates the same rules many times — the 5-10 paper floor, 'why now / what changed', bounded divergence to a 2-3 frontier, and selection-gate checks each recur across 'Control workflow', 'Creative-divergence protocol', 'Integrated ideation workflow', 'Workflow', and 'Non-negotiable rules', which is padded with redundant context rather than lean guidance.	1 / 3
Actionability	Guidance is concrete and executable: specific bundle references (references/selection-gate.md, references/literature-survey-template.md), durable artifact paths (artifacts/idea/selected_idea.md), explicit API calls (artifact.submit_idea(...), memory.search(...), artifact.arxiv(paper_id=..., full_text=False)), and numeric thresholds (6-12 raw ideas, 2-3 frontier, 7/10 gate).	3 / 3
Workflow Clarity	Workflows are sequenced with explicit validation checkpoints and feedback loops: the 7.1 quality gate ('If the total is below 7/10, do not promote'), pre-idea-draft challenge before submission, and exit criteria ('Do not exit this stage with a selected idea if the literature survey report is missing...') with route-back-to-decision/scout recovery.	3 / 3
Progressive Disclosure	References are real, one level deep, and clearly signaled (all 13 referenced references/*.md files exist), but the SKILL.md itself is a monolithic ~1500-line document with substantial detail kept inline that could be split into references, so 'content appropriately split' is not met.	2 / 3
	Total	9 / 12 Passed

Description

50%

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 concise, uses third person, and includes an explicit 'Use when' trigger with several named deliverables, but it lacks an explicit action verb stating what the skill does, leaving the 'what' implied. It is competent but not exemplary, landing at the midpoint across all dimensions.

Suggestions

Lead with an explicit third-person action verb before the trigger, e.g. 'Generates literature-grounded candidate research directions and selects a falsifiable next route. Use when...'.

Add a few more natural trigger phrasings users would actually say ('brainstorm research ideas', 'decide what to try next', 'find a new direction after a failed line') alongside the existing deliverable terms.

Sharpen distinctiveness from sibling skills by naming the boundary, e.g. 'Use for direction selection, not literature expansion (scout) or within-family tuning (optimize).'

Dimension	Reasoning	Score
Specificity	The description names the domain and several concrete deliverables ("concrete hypotheses, limitation analysis, candidate directions, or a selected idea"), but it frames them as what the quest 'needs' rather than stating action verbs the skill performs, so it does not reach the 'lists multiple specific concrete actions' anchor.	2 / 3
Completeness	It has an explicit 'Use when' trigger (the 'when'), but the 'what does this do' is only implied via the deliverables the quest needs; per the judging guideline not to infer, the missing explicit action verb keeps it below the both-what-and-when anchor.	2 / 3
Trigger Term Quality	It includes relevant terms a research-agent orchestrator might say ("concrete hypotheses", "candidate directions", "selected idea", "active baseline"), but several are system-internal jargon and it misses common variations like 'brainstorm research ideas' or 'what to try next'.	2 / 3
Distinctiveness Conflict Risk	The niche ('idea' / direction selection relative to an active baseline) is fairly specific, but it overlaps noticeably with sibling skills like 'scout' and 'optimize' — a risk the body itself spends considerable effort disambiguating.	2 / 3
	Total	8 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1500 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: ResearAI/DeepScientist
Path: src/skills/idea/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.