pulser-eval
Run cross-model evals on the Pulser for Odoo platform. Requires explicit resource_config documentation to ensure comparable results.
来源信息
- 仓库
- Insightpulseai/odoo
- 最近来源活动
- 2026年4月17日 17:39
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 6
- 分支
- 2
安装方式
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
检查来源文件
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
正在显示 SKILL.md
SKILL.md
来源说明 · 只读预览- name
- pulser-eval
- description
- Run cross-model evals on the Pulser for Odoo platform. Requires explicit resource_config documentation to ensure comparable results.
- disable-model-invocation
- true
- user-invocable
- true
- allowed-tools
- Bash(python3 scripts/run_eval.py)
- inputs
- [{"dataset":"String - path to the eval dataset."},{"resource_config":"String (e.g., 'high-headroom-v1') - The ACA resource footprint for this eval."},{"odoo_version":"String (e.g., '18.0')"}]
# pulser-eval
This skill executes high-fidelity evaluations against Odoo 18 CE environments.
## Eval Infrastructure Rule (Anthropic Engineering Blog)
Eval scores are not comparable across different infrastructure configurations.
A 3x ceiling over per-task baseline specs is required to eliminate OOM-skew.
Registration of \`resource_config\` is mandatory.
## Usage
- Trigger: "run evals on the AP flow"
- Required metadata: \`resource_config\`
在 GitHub 查看