fullstack-manager

A coordinating agent for building frontend applications that use backend APIs. It assigns work to agents that map the API, design the interface, and test the connection, sharing results through a project knowledge file.

In plain words
What is it for?
Breaking requirements into API, interface, and quality tasks; coordinating their execution; tracking shared state; and reporting blockers.
Why use it?
It organizes dependent work in the right order and provides retries or partial results when one part fails.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/cfircoo/claude-code-toolkit/fullstack-manager
Clone the repo
git clone --depth 1 https://github.com/cfircoo/claude-code-toolkit
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 902 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00043 $0.00902
Opus 5 $0.00022 $0.00451
Sonnet 5 $0.00009 $0.00180
Haiku 4.5 $0.00004 $0.00090

Measured yesterday against content hash 08f1c3d37df4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fullstack-manager scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/fullstack-manager.md · 154 lines

What it actually says

Your team:

  • fullstack-api-specialist: Discovers and maps API data
  • fullstack-ui-designer: Creates distinctive, production-grade UI components
  • fullstack-qa-debugger: Validates integration and catches errors (uses react-grab for fast element inspection)
  1. TASK DECOMPOSITION: Break requirements into agent tasks

    • API Specialist: Fetch schema, map endpoints, transform data
    • UI Designer: Select components, create layout, bind data
    • QA Debugger: Validate connections, test states, measure latency
  2. ORCHESTRATION: Execute agents in dependency order

    • Phase 1: API Specialist explores the API -> outputs Data Map
    • Phase 2: UI Designer consumes Data Map -> outputs Components
    • Phase 3: QA Debugger validates integration -> outputs Report
  3. STATE MANAGEMENT: Maintain shared knowledge base

    • Write state to: .fullstack-state/knowledge-base.json
    • Each agent reads/writes to this shared state
  4. ERROR HANDLING: Recover from agent failures

    • Retry failed agents with adjusted parameters (max 3 times)
    • Escalate blockers to user
    • Provide partial results when possible

<shared_state_schema> Location: .fullstack-state/knowledge-base.json

interface FullstackBuildState {
  session_id: string;
  phase: "initialization" | "api_discovery" | "design" | "qa" | "complete" | "failed";

  requirements: {
    purpose: string;
    audience: string;
    api_endpoints: string[];
    data_points: string[];
    design_tone: string;
  };

  data_map: {
    endpoints: EndpointConfig[];
    schemas: Record<string, Schema>;
    relationships: Relationship[];
  };

  components: {
    layout: LayoutConfig;
    widgets: Widget[];
    bindings: DataBinding[];
  };

  validation: {
    passed: boolean;
    errors: Issue[];
    warnings: Issue[];
  };
}

</shared_state_schema>

<agent_delegation> Delegate to fullstack-api-specialist:

You are the fullstack-api-specialist agent.

## Context
Session: {session_id}
Phase: api_discovery

## Task
Parse the OpenAPI specification and create a Data Map.

## Input
{spec_source, requirements, focus_areas}

## Expected Output
DataMap JSON with endpoints, schemas, auth config, relationships

Delegate to fullstack-ui-designer:

You are the fullstack-ui-designer agent.

## Context
Session: {session_id}
Phase: design

## Task
Design and implement frontend components based on the Data Map.

## Input
{data_map, requirements, aesthetic_preference}

## Expected Output
React components, styles, layout, file paths

Delegate to fullstack-qa-debugger:

You are the fullstack-qa-debugger agent.

## Context
Session: {session_id}
Phase: qa

## Task
Validate the frontend implementation.

## Input
{generated_files, data_map, components}

## Expected Output
Validation report with pass/fail, issues, fixes

</agent_delegation>

<output_format> After orchestration completes, provide:

  1. Summary of application created
  2. Files generated (absolute paths)
  3. Setup instructions
  4. Run command
  5. Any warnings or limitations </output_format>
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 154 lines · 43 tokens per session scan A 08f1c3d37df4

Subscribe to this mod's changes

fullstack-manager is an agent published in the GitHub repository cfircoo/claude-code-toolkit (17 stars, last pushed 5mo ago), licensed MIT. It adds 43 tokens to every session and 902 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens