honeybadge/harbor
Framework for AI agent evaluation in containerized environments. Use when: (1) Running agent evaluations with `harbor run` against benchmarks (SWE-Bench, Terminal-Bench, Aider Polyglot, etc.), (2) Creating custom benchmark tasks with Dockerfile, instruction.md, solution, and tests, (3) Building adapters to convert existing benchmarks to Harbor format, (4) Implementing custom agents extending BaseAgent or BaseInstalledAgent, (5) Scaling evaluations to cloud providers (Daytona, Modal, E2B), (6) Exporting traces for RL/SFT training, (7) Debugging Harbor runs or inspecting package internals.
99
Does it follow best practices?
Validation for skill structure
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 an excellent skill description that clearly defines a specialized domain (AI agent evaluation in containerized environments) with comprehensive trigger scenarios. It uses third person voice, provides specific tool and benchmark names, and explicitly lists seven distinct use cases that would help Claude accurately select this skill.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: running evaluations with `harbor run`, creating custom benchmark tasks, building adapters, implementing custom agents, scaling to cloud providers, exporting traces, and debugging runs. | 3 / 3 |
Completeness | Clearly answers both what (framework for AI agent evaluation in containerized environments) and when (explicit numbered list of 7 specific trigger scenarios with 'Use when:' clause). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'harbor run', 'SWE-Bench', 'Terminal-Bench', 'Aider Polyglot', 'benchmark', 'Dockerfile', 'BaseAgent', 'Daytona', 'Modal', 'E2B', 'RL/SFT training', 'traces'. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with specific tool names (Harbor, SWE-Bench, Terminal-Bench), class names (BaseAgent, BaseInstalledAgent), and cloud providers (Daytona, Modal, E2B) that create a clear niche unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an exemplary skill file that maximizes information density while maintaining clarity. It uses tables effectively for reference material, provides executable examples throughout, and structures content with clear progressive disclosure to detailed guides. The troubleshooting workflow demonstrates proper validation checkpoints.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely lean and efficient. Uses tables for dense information, no unnecessary explanations of concepts Claude would know. Every section earns its place with actionable content. | 3 / 3 |
Actionability | Provides fully executable commands throughout - install, run benchmarks, list datasets, scale to cloud. The CLI examples are copy-paste ready with real flags and model names. | 3 / 3 |
Workflow Clarity | Troubleshooting section provides clear numbered workflow with explicit checkpoints (check logs → verify Dockerfile → increase timeout → resume → debug mode). The progression from failure to resolution is well-sequenced. | 3 / 3 |
Progressive Disclosure | Excellent structure with quick start overview, then detailed guides table pointing to one-level-deep references (tasks.md, adapters.md, agents-usage.md, commands.md). Clear signaling of when to read each guide. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
93%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 |
|---|---|---|
metadata_version | 'metadata' field is not a dictionary | Warning |
Total | 15 / 16 Passed | |
Install with Tessl CLI
npx tessl i honeybadge/harborReviewed
Table of Contents