convention-injector

Inject task-specific conventions into subagent prompts to govern scope.

15|6|Updated May 7, 2026

Search Tags:#automation #subagents #conventions #claude.md #rules #prompt-injection #anti-contamination

Ready to run

npx skills add https://github.com/MinhThang1009/dotclaude --skill convention-injector

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: convention-injector
Source: https://github.com/MinhThang1009/dotclaude/tree/main/plugins/subagent-system/skills/convention-injector
Command: npx skills add https://github.com/MinhThang1009/dotclaude --skill convention-injector

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill ensures that subagents operate with task-specific conventions, preventing contamination and ensuring consistent prompt governance across Explore/Plan agents.

Core Features & Use Cases

Collects conventions from user and project CLAUDE.md and rules/*.md.
Filters and injects only relevant conventions (≤300 tokens) into subagent prompts.
Provides an anti-contamination option for audit-type tasks and adds a self-confirmation prompt.

Quick Start

Before spawning a subagent, run convention-injector to inject task-specific conventions into its prompt.

Frequently Asked Questions about convention-injector

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I inject task-specific conventions into subagents to prevent prompt contamination?▼

To inject task-specific conventions into subagents and prevent prompt contamination, use a skill that reads user and project CLAUDE.md files, filters relevant rules, and appends a compact conventions block with a self-confirmation suffix to the subagent prompt.

What is prompt contamination in subagents and how do CLAUDE.md rules prevent it?▼

Prompt contamination occurs when subagents execute tasks with irrelevant or conflicting instructions. CLAUDE.md rules prevent it by enabling task-specific filtering, ensuring subagents only receive applicable conventions within a strict token limit.

How do I apply project conventions when spawning Explore and Plan agents?▼

To apply project conventions when spawning Explore and Plan agents, run the convention injection process before spawning to load CLAUDE.md, filter relevant rules, and inject a compact conventions block into the agent prompt.

Does convention injection work with user-level and project-level rules simultaneously?▼

Yes, convention injection works with user-level and project-level rules simultaneously by collecting conventions from both user and project CLAUDE.md files, as well as rules/*.md, to filter and apply relevant rules to the subagent.

What is the token limit for injected conventions in subagent prompts?▼

The token limit for injected conventions in subagent prompts is 300 tokens. The injection mechanism filters relevant rules to create a compact conventions block that stays within this strict limit to prevent prompt bloat.

How do I add a self-confirmation prompt for audit-type subagent tasks?▼

To add a self-confirmation prompt for audit-type subagent tasks, use the anti-contamination option provided by the convention injector, which appends a self-confirmation suffix to the subagent prompt to verify rule adherence.