experiment-bridge

Workflow 1.5: Bridge between idea discovery and auto review. Reads EXPERIMENT_PLAN.md, implements experiment code, deploys to GPU, collects initial results. Use when user says "实现实验", "implement experiments", "bridge", "从计划到跑实验", "deploy the plan", or has an experiment plan ready to execute.

Quality

88%

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

81%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 well-sequenced, actionable runbook with strong validation and feedback loops, scoring high on workflow clarity. Its main weaknesses are a boilerplate-heavy inlined review prompt and external/possibly-dangling reference paths with no local bundle files.

Suggestions

Extract the Codex review prompt (especially the generic SCOPE LIMITS preamble) into a referenced file and inline only the skill-specific review checklist to tighten conciseness.

Move the failure-classification/debugging taxonomy in Phase 3 into a shared-references file, keeping only the budget rule and the /codex:rescue pointer inline.

Verify the referenced paths (shared-references/*.md, templates/RESEARCH_CONTRACT_TEMPLATE.md, skills/experiment-queue/SKILL.md) resolve from the skill's installed location, or mark them explicitly as suite-relative.

Dimension	Reasoning	Score
Conciseness	Largely lean and operational with no concept padding, but the inlined Codex review prompt includes a sizable "SCOPE LIMITS" boilerplate block whose generic review-guidance text does not earn its place for this specific skill.	4 / 5
Actionability	Provides concrete executable commands (git clone, /run-experiment, /experiment-queue, train.py invocations) and a precise auto-routing rule, though some blocks remain fill-in templates and core script contents are delegated to the plan.	4 / 5
Workflow Clarity	Six sequenced phases with explicit validation checkpoints (sanity gate, CRITICAL-issue review gate) and a validate→fix→retry→reimplement feedback loop with a bounded budget — the validation/verification the batch-operation cap requires is present.	5 / 5
Progressive Disclosure	Well-organized into clear sections with clearly signaled one-level-deep references to shared-references and templates, but no bundle files exist in references/scripts/assets and several referenced paths are external/possibly dangling, with one large prompt block inlined that could live in a reference file.	4 / 5
	Total	17 / 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 strong: it concisely states four concrete actions and gives an explicit, bilingual Use-when trigger clause. The only weakness is minor overlap risk from the generic term "bridge" and adjacency to sibling experiment skills.

Suggestions

Consider dropping or qualifying the standalone trigger "bridge" (e.g., "experiment bridge") to reduce overlap with unrelated bridging skills.

Add EXPERIMENT_PLAN.md as an explicit trigger token so users who mention the file by name route correctly.

Dimension	Reasoning	Score
Specificity	Names four concrete actions ("Reads EXPERIMENT_PLAN.md, implements experiment code, deploys to GPU, collects initial results") with comprehensive coverage of the skill's scope, matching the anchor for multiple specific concrete actions.	5 / 5
Completeness	Explicitly answers both what (read/implement/deploy/collect) and when ("Use when user says ...") with concrete trigger phrases, matching the top anchor exactly.	5 / 5
Trigger Term Quality	Provides six explicit bilingual trigger phrases ("实现实验", "implement experiments", "bridge", "从计划到跑实验", "deploy the plan") plus a state-based trigger, giving comprehensive synonym coverage of natural phrasing.	5 / 5
Distinctiveness Conflict Risk	Has a clear experiment-execution niche with distinct triggers, but "bridge" alone is generic and "implement experiments" carries minor overlap risk with sibling experiment skills in the same suite.	4 / 5
	Total	19 / 20 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 3 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/experiment-bridge/SKILL.md
Commit: f4f20f9

Reviewed: 1 day 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.