Back to Skills

context-assembler

Assembles relevant context for agent spawns with prioritized ranking. Ranks packages by relevance, enforces token budgets with graduated zones, captures error patterns for learning, and supports configurable per-agent retrieval limits.

20starsUpdated 3/25/2026

Security Assessment

Critical Risk(0/100)

Detected risks:

Remote Code Execution([SKILL.md] eval ()
Secret Exposure([SKILL.md] TOKEN =, [SKILL.md] TOKEN=)
Command Injection([SKILL.md] subprocess.run)
Security Score0/100

About context-assembler

The context-assembler skill is designed to assemble and prioritize relevant context packages for agent spawns, ensuring the agent has access to the necessary information while adhering to token budget constraints. This skill is essential in orchestration systems that require context-rich input for various types of agents. It helps streamline the process of preparing agents by dynamically selecting context packages that are most pertinent to the task at hand, allowing for more efficient and accurate agent performance.

This skill is equipped with several key features, such as ranking context packages based on relevance, enforcing token budgets with graduated zones, and capturing error patterns for learning. It is also highly configurable, allowing users to set specific retrieval limits per agent type. Additionally, the context-assembler skill can be invoked automatically based on agent types and session contexts, making it a highly adaptive tool for a variety of orchestration scenarios. Its primary functionality is ensuring that agents such as developers, QA experts, and tech leads receive the right information based on their role and the stage of the task.

Use cases for the context-assembler skill include preparing agents for developer or QA tasks, ensuring they have the context they need for their specific role, and facilitating the debugging and reasoning processes by providing full context to agents like investigators. It is ideal for environments where agents need tailored, context-specific information for their work, making it an indispensable tool in agent orchestration workflows.

FAQ

When should I invoke the context-assembler skill?

Invoke this skill when the orchestrator is preparing to spawn an agent and needs relevant context, or when an agent requests context assembly during a task. It is also needed for error pattern capture and prior context retrieval before spawning certain agents.

What are the limitations of the context-assembler skill?

The context-assembler skill should not be invoked outside the BAZINGA orchestration context. It also does not support manual reading of specific files—use the Read tool for that.

Can I configure the retrieval limit for different agent types?

Yes, the retrieval limits for agent types like developer, QA expert, and tech lead can be configured in the skill's configuration. The default limits are 3 for developers and 5 for other roles, but these can be adjusted as needed.

Does the context-assembler skill support token budgeting?

Yes, the context-assembler skill supports token budgeting through graduated zones. It estimates token usage and ensures that the context packages provided do not exceed the available budget for the given agent.

Is FTS5 required for context ranking?

No, FTS5 is not required. The context-assembler skill uses a heuristic fallback for context ranking when FTS5 is unavailable.

Install context-assembler

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill

Repository

mehdic/bazinga