product-grounded-artifact-retrieval
Load enterprise data from /root/DATA, identify the correct artifact version for each product mentioned in questions, and reject cross-product distractors. Apply strict 2-signal product grounding (artifact metadata + question context).
- License 许可
- MIT
- 最近更新
- 2026-10-02
- Stars
- ★ 93
- Forks
- 8
01 前言 Front Matter
- name
- product-grounded-artifact-retrieval
- description
- Load enterprise data from /root/DATA, identify the correct artifact version for each product mentioned in questions, and reject cross-product distractors. Apply strict 2-signal product grounding (artifact metadata + question context).
03 SKILL.md 正文
Product-Grounded Artifact Retrieval Skill
Step 1: Load Enterprise Data
- Read all files from
/root/DATAdirectory - Index artifacts by product name and version (draft vs. final)
- Track artifact metadata: creation date, status, product association
- Build a product-to-artifacts mapping
Step 2: Two-Signal Product Grounding
For each question's artifact context:
- Signal 1: Match product name from question to artifact metadata
- Signal 2: Verify artifact status/version (prefer final versions unless explicitly asking for draft)
- Reject any artifacts from different products, even if semantically similar
- Reject draft versions unless the question specifically asks for drafts
- Document which artifacts were rejected and why
Step 3: Validate Product Isolation
- Ensure no cross-product confusion (e.g., Product A reviewers mixed with Product B)
- If multiple artifact versions exist for a product, select the correct one based on question intent
- Fail explicitly if product grounding is ambiguous
Step 4: Return Grounded Artifacts
Return a mapping of:
{
"question_id": "q1",
"artifact_id": "artifact_name",
"product": "product_name",
"version": "final|draft",
"valid": true/false,
"rejection_reason": "if not valid"
}
04 同仓库更多 Skills
object_counter
Count occurrences of an object in the image using computer vision algorithm.
travel-planning-workflow
Invoke this skill when performing travel planning. It defines a structured workflow for generating itineraries. Following this workflow is essential to ensure the final plan satisfies user requirements.
bundled-dataset-search-tool
invoke this skill when you need to perform database search for travel planning. This skill provides some useful pre-packaged tools to look up accommodations, attractions, cities, driving distance, flights, and restaurants from the bundled dataset.
glm-output
Output-processing guidance for GLM tasks. Especially useful after glm-basics and glm-calibration when you need verifier-matching metrics from output.nc and a final /root/metrics.json.
glm-calibration
Calibration guidance for GLM tasks. Often most effective after glm-basics has clarified the setup; glm-output is the companion skill for exact final metric computation.
glm-basics
A good starting point for GLM calibration tasks. Use to inspect glm3.nml, confirm how GLM runs, and identify the relevant files before moving on to calibration and output evaluation.