idea-discovery-robot

Workflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware literature survey, idea generation, novelty check, and critical review to go from a broad robotics direction to benchmark-grounded, simulation-first ideas. Use when user says \"robotics idea discovery\", \"机器人找idea\", \"embodied AI idea\", \"机器人方向探索\", \"sim2real 选题\", or wants ideas for manipulation, locomotion, navigation, drones, humanoids, or general robot learning.

Quality

93%

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

Quality

Content

88%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-engineered orchestration skill: every phase has an exact sub-skill invocation with concrete arguments, checkpoints with defined fallback behavior, and safety gating on real-robot execution. The main improvement opportunities are trimming rhetorical and repeated guidance, and moving bulky templates/matrices into clearly signaled reference files.

Suggestions

Consolidate the overlapping guidance in Good/Weak Idea Patterns, Filtering Rules, and Key Rules into a single evaluation-criteria section to remove repetition and reduce token load.

Move the Phase 6 report template and the Phase 1 landscape-matrix axes into a references/ file (e.g. REPORT_TEMPLATE.md) linked from the body, slimming SKILL.md to the orchestration logic.

Drop rhetorical framing like 'The goal is not to produce flashy demos. The goal is to produce ideas that are...' in favor of the equivalent bullet list alone.

Dimension	Reasoning	Score
Conciseness	The body is efficient instruction prose with tables, templates, and exact sub-skill prompts, and it does not explain concepts Claude already knows. However, there is minor rhetoric and repetition — e.g. "The goal is not to produce flashy demos. The goal is to produce ideas that are..." and overlapping guidance across Good/Weak Idea Patterns, Filtering Rules, and Key Rules — matching the score-4 anchor (minor instances that could be trimmed) rather than the every-token-earns-its-place score-5 anchor.	4 / 5
Actionability	It provides copy-paste-ready invocations with full argument strings for every sub-skill (/research-lit, /idea-creator with the robotics frame pasted in, /novelty-check with the required axes, /research-review with reviewer framing), plus a landscape-matrix table, a pilot-spec template, and a complete report template — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	Phases 0–6 are strictly ordered with an Execution Rule, AUTO_PROCEED fallback behavior, explicit checkpoints after Phases 1 and 2, recovery paths on user feedback (refine frame, re-run phase), and a hard Real Robot Rule gating hardware execution — clear sequencing with explicit validation checkpoints and feedback loops.	5 / 5
Progressive Disclosure	The single-file skill is well-sectioned per phase, and its external references (../../shared-references/output-versioning.md, output-manifest.md, output-language.md) are clearly signaled and one level deep. It falls short of the score-5 anchor because those shared references live outside the skill directory (unverifiable in the bundle), and some inline material — the full report template and landscape matrix — could be split into reference files to slim the 360-line body.	4 / 5
	Total	18 / 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.

A strong description: it states a concrete multi-stage pipeline, a clear outcome, and an explicit 'Use when' clause with rich bilingual trigger phrases. Its only weaknesses are the unexplained 'Workflow 1 adaptation' jargon and slight breadth at the tail of the trigger list.

Suggestions

Replace the opaque 'Workflow 1 adaptation for robotics and embodied AI' lead-in with a plain capability statement, e.g. 'Runs a robotics idea-discovery pipeline'.

Clarify how this skill differs from the generic idea-discovery workflow it adapts (e.g. 'Use this robotics variant instead of generic idea discovery when the direction involves robots, simulators, or embodied AI') to reduce sibling-skill overlap.

Dimension	Reasoning	Score
Specificity	The description lists four concrete pipeline actions — "robotics-aware literature survey, idea generation, novelty check, and critical review" — plus a stated outcome ("benchmark-grounded, simulation-first ideas"), giving comprehensive concrete coverage rather than the minor gaps of the score-4 anchor.	5 / 5
Completeness	It explicitly answers both what ("orchestrates ... literature survey, idea generation, novelty check, and critical review") and when ("Use when user says ..." with concrete trigger phrases), matching the score-5 anchor.	5 / 5
Trigger Term Quality	Trigger coverage is comprehensive and natural, including synonyms and bilingual variants: "robotics idea discovery", "embodied AI idea", "sim2real 选题", "机器人找idea", plus topic-level terms (manipulation, locomotion, navigation, drones, humanoids, robot learning).	5 / 5
Distinctiveness Conflict Risk	The robotics/embodied-AI niche and triggers are largely distinct, but "Workflow 1 adaptation" implies a closely related generic idea-discovery skill, and "general robot learning" is broad, leaving minor overlap risk with that sibling — the score-4 anchor rather than the clear-niche score-5 anchor.	4 / 5
	Total	19 / 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 suspicious	Warning

	Total	15 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/skills-codex-gemini-review/idea-discovery-robot/SKILL.md
Commit: 26b95cf

Reviewed: 8 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.