Documents conditional context block assembly rules for LLM prompt nodes.
npx skills add https://github.com/wlee075/chatbot --skill context-assembly-wlee075
Please help me install this Agent Skill. Skill: context-assembly Source: https://github.com/wlee075/chatbot/tree/main/skills/context_assembly Command: npx skills add https://github.com/wlee075/chatbot --skill context-assembly-wlee075
Context-heavy prompt systems across multiple LLM nodes often struggle with maintaining consistent, auditable context blocks. This skill documents the precise rules for which blocks are assembled and injected, preventing duplication, conflicts, and leakage of irrelevant information.
Review the documented assembly rules and apply them to your LLM prompt system to ensure consistent, auditable context injection across nodes.
High-intent search queries and answers about installing and using this skill.
You can conditionally assemble LLM prompt context blocks across multiple nodes by applying documented block-injection rules, which prevent duplication, conflicts, and irrelevant information leakage. Guard rules ensure only relevant context compiles per section and iteration.
The best way to maintain consistent context injection is applying shared block templates and format contracts across Elicitor, Drafter, and Reflector nodes. This auditable workflow ensures reliable behavior and simplifies change management for context-heavy prompt systems.
The _format_prd_so_far helper renders the accumulated PRD context into the final prompt block. It works alongside core templates like ELICITOR_CONTEXT_BLOCK and ELICITOR_PRD_BLOCK to conditionally inject relevant formatted context into the active LLM node.
You need conditional block rendering when your prompt system spans multiple LLM nodes and struggles with maintaining consistent, auditable context. It compiles only relevant blocks based on section, iteration, and state to prevent context conflicts and information leakage.
Yes, the block-injection rules explicitly apply to Elicitor, Drafter, and Reflector nodes to maintain consistent behavior. Any multi-node LLM workflow requiring auditable context blocks and format contracts can adapt these conditional assembly rules.
EmeaAppGbbBuild and orchestrate AI agents with Microsoft Agent Framework.
Probably-GroupCreates secure, rule-based prompts for LLM workflows with structured outputs.
CoworkedShawnDetect and block prompt injection attempts in OpenClaw agents.
thejefflarsonDetect token-smuggling and Unicode injection patterns in LLM prompts.
