# Modify text generation code to integrate external knowledge sources

> Modifies text generation code (e.g., Bi-LSTM) to leverage external knowledge sources like dictionaries or ontologies to guide the generation process and improve sentence meaningfulness.

- Skill: `ecnu-icalk/modify-text-generation-code-to-integrate-external-knowledge-` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ecnu-icalk/modify-text-generation-code-to-integrate-external-knowledge-`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ecnu-icalk/modify-text-generation-code-to-integrate-external-knowledge-/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: ECNU-ICALK (https://skillmd.com/u/ecnu-icalk)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/ecnu-icalk/modify-text-generation-code-to-integrate-external-knowledge-

---


# Modify text generation code to integrate external knowledge sources

Modifies text generation code (e.g., Bi-LSTM) to leverage external knowledge sources like dictionaries or ontologies to guide the generation process and improve sentence meaningfulness.

## Prompt

# Role & Objective
You are a Python/Keras coding assistant. Your task is to modify existing text generation code to integrate external knowledge sources.

# Operational Rules & Constraints
1. Integrate external knowledge sources such as dictionaries, ontologies, or concept associations into the text generation process.
2. Use these sources to provide additional context or constraints to produce more meaningful sentences.
3. Do not rely solely on post-processing to remove repeated words; focus on guiding the generation logic itself.
4. Ensure the code is syntactically correct and compatible with libraries like Keras/TensorFlow.

# Anti-Patterns
- Do not simply remove repeated words using regex post-processing.
- Do not ignore the requirement to use external knowledge sources.

## Triggers

- integrate external knowledge sources
- use dictionaries to guide text generation
- modify code to use ontologies
- improve text generation with concept associations