fibonacci-recursive-agents

Computes Fibonacci numbers through recursive agent delegation where each agent spawns child agents for sub-problems.

36|5|Updated Feb 27, 2026
One-click install
npx skills add https://github.com/Birdhouse-Labs/birdhouse --skill fibonacci-recursive-agents-birdhouse-labs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fibonacci-recursive-agents
Source: https://github.com/Birdhouse-Labs/birdhouse/tree/main/.agents/skills/internal/test-cases/fibonacci-recursive-agents
Command: npx skills add https://github.com/Birdhouse-Labs/birdhouse --skill fibonacci-recursive-agents-birdhouse-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Testing recursive multi-agent delegation requires a verifiable workload, and manually orchestrating nested agent trees is error-prone. This Skill provides a deterministic test case where Fibonacci computation naturally produces an exponentially growing agent tree. ## Core Features & Use Cases - Recursive Agent Spawning: Each agent handling fib(N) with N >= 2 must spawn two child agents for fib(N-1) and fib(N-2), never computing sub-problems itself. - Agent Tree Rendering: After the root agent completes, the Skill calls agent_tree() and renders the full hierarchy as nested markdown links with birdhouse:agent/ URIs. - Use Case: Run a fibonacci test with default n=3 to verify that the Birdhouse agent runtime correctly creates, tracks, and reports parent-child agent relationships across multiple recursion levels. ## Quick Start Ask the agent to run a fibonacci recursive test with n=3 and render the resulting agent tree.

Frequently Asked Questions about fibonacci-recursive-agents

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

FAQPage Schema
How do I run a recursive Fibonacci agent test?▼

Invoke the Skill with a phrase like "run fibonacci test" and optionally specify n (default is 3). The Skill creates a root agent with the recursive prompt, waits for completion, then renders the full agent tree as nested markdown links.

How does recursive agent delegation work for Fibonacci?▼

Each agent receiving fib(N) with N >= 2 spawns two child agents for fib(N-1) and fib(N-2) instead of computing them directly. Base cases fib(0) and fib(1) return immediately without children, and parents sum their children's results.

What does the agent tree output look like?▼

The output shows the computed result followed by a nested markdown list where each node is a link like [fib(N)](birdhouse:agent/agent_ID). Indentation of two spaces per level reflects the parent-child hierarchy from the root agent downward.

Why does the agent tree grow exponentially with larger n?▼

Naive recursive Fibonacci spawns two children per non-base call, so the agent count follows the Fibonacci recurrence itself. For n=3 the tree has 5 agents, but larger n values create exponentially more agents, which is the point of the stress test.

When should I not use this recursive Fibonacci test?▼

Avoid large n values since agent count grows exponentially and can exhaust runtime resources. This Skill is an internal test case for validating agent delegation, not an efficient way to compute Fibonacci numbers.