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.
88
85%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that clearly defines a specific workflow step with concrete actions, explicit trigger terms in multiple languages, and a well-defined 'Use when' clause. The description effectively positions itself within a larger workflow pipeline, making it highly distinctive and easy for Claude to select appropriately.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: reads EXPERIMENT_PLAN.md, implements experiment code, deploys to GPU, collects initial results. These are clear, actionable steps in a defined workflow. | 3 / 3 |
Completeness | Clearly answers both 'what' (reads experiment plan, implements code, deploys to GPU, collects results) and 'when' (explicit 'Use when' clause with specific trigger phrases and a situational trigger). Both dimensions are well-covered. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms in both English and Chinese: '实现实验', 'implement experiments', 'bridge', '从计划到跑实验', 'deploy the plan', plus the contextual trigger 'experiment plan ready to execute'. Good coverage of how users would naturally phrase this request. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: it's specifically a bridge workflow (1.5) between idea discovery and auto review, targeting GPU deployment of experiment plans. The specific file reference (EXPERIMENT_PLAN.md) and workflow positioning make it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured orchestration skill with excellent workflow clarity and progressive disclosure. Its main weakness is that actionable guidance is more template-oriented than executable — commands use placeholders rather than concrete examples, and the implementation phase (Phase 2) describes what to do rather than showing how. The skill is moderately concise for its complexity but could trim some of the verbose status message templates.
Suggestions
Replace placeholder commands like `/run-experiment [sanity experiment command]` with at least one concrete, realistic example showing actual arguments and expected output.
In Phase 2, add a small concrete code snippet showing what a properly implemented experiment script looks like (e.g., argparse setup with seed, metric saving to JSON) rather than just listing requirements.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly long (~200 lines) with some verbosity in template outputs and status messages, but most content is genuinely instructive. The overview diagram and composing section add value. Some sections like the constants explanations could be tighter, and the multiple template blocks add bulk, but overall it's reasonably efficient for a complex multi-phase workflow. | 2 / 3 |
Actionability | The skill provides clear step-by-step phases with specific file paths, tool invocations (/run-experiment, /monitor-experiment), and output templates. However, actual executable code is minimal — most commands are placeholders like `/run-experiment [sanity experiment command]` rather than concrete examples. The self-review checklist in Phase 2 is actionable, but implementation guidance is more descriptive than executable. | 2 / 3 |
Workflow Clarity | Excellent multi-phase workflow with clear sequencing (parse → implement → sanity check → deploy → collect → handoff). Includes explicit validation checkpoints: sanity check before full deployment, self-review checklist before deploying code, AUTO_DEPLOY checkpoint, success criteria checking, and a 'fix and re-run' feedback loop for sanity failures. The milestone ordering is well-defined with clear gates between phases. | 3 / 3 |
Progressive Disclosure | Content is well-structured with clear sections and phases. References to other skills (/run-experiment, /monitor-experiment, /auto-review-loop, /ablation-planner, /training-check) are one level deep and clearly signaled. The composing section at the end provides excellent navigation context. Input files are clearly listed with priority ordering. For a skill of this complexity, the organization is appropriate. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
100%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 / 11 Passed
Validation for skill structure
No warnings or errors.
- Commit
dc00dfb
- 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.