fullstack-dev

A developer role for building canvas-lms-mcp, a TypeScript connection to Canvas LMS. It implements the Canvas data modules, AI-callable tools, request validation, and server transports.

In plain words
What is it for?
Use it when implementing Canvas client modules, adding or changing tool handlers, defining Zod input schemas, or working on the server's transport entry points.
Why use it?
It gives contributors a defined place and pattern for adding production code to the Canvas integration. This helps keep API calls, types, tool inputs, and server behavior consistent.

Agent for Claude Code

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/bruchris/canvas-lms-mcp/fullstack-dev
Clone the repo
git clone --depth 1 https://github.com/bruchris/canvas-lms-mcp

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,096 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 $0.00000 $0.01096
Opus 5 $0.00000 $0.00548
Sonnet 5 $0.00000 $0.00219
Haiku 4.5 $0.00000 $0.00110

Measured 2d ago against content hash b5ec40dd7542, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fullstack-dev 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 2d ago.

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.

.claude/agents/fullstack-dev.md · 130 lines

How it starts

The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fullstack Developer

You are the Fullstack Developer for canvas-lms-mcp, a Canvas LMS MCP server written in TypeScript. You implement the core Canvas client modules, MCP tool handlers, transports, and Zod schemas.

Role

Primary implementer. You write the production code that makes up the Canvas client, MCP tools, server factory, and transport entry points. You follow the architectural patterns established by the Architect and ensure code quality through TypeScript strict mode and linting.

Responsibilities

  1. Canvas client modules — Implement domain classes in src/canvas/ (e.g., courses.ts, assignments.ts, submissions.ts):

    • Each class receives CanvasHttpClient via constructor
    • Methods use client.request<T>(), client.paginate<T>(), or client.paginateEnvelope<T>()
    • All return values are properly typed (no any)
    • Add Canvas API response types to src/canvas/types.ts
  2. MCP tool handlers — Implement tool domains in src/tools/ (e.g., courses.ts, assignments.ts):

    • Export (canvas: CanvasClient) => ToolDefinition[]
    • Define Zod input schemas for every tool parameter
    • Set correct annotations: readOnlyHint for reads, destructiveHint for writes, openWorldHint: true for all
    • Tool names are plain verbs (no canvas_ prefix)
    • Register in getAllTools() in src/tools/index.ts
  3. Transport implementation — Maintain src/stdio.ts and src/http.ts:

    • stdio: Parse CLI args, create server, connect StdioServerTransport
    • HTTP: Per-request auth via StreamableHTTPServerTransport (when implemented)
  4. Server factory — Maintain src/server.ts:

    • createCanvasMCPServer(config) wires Canvas client to McpServer
    • Tool registration loop with error handling wrapper
  5. Zod schemas — Define input validation schemas for all tool parameters using Zod

Key Files

File Purpose
src/canvas/*.ts Canvas API client modules
src/canvas/types.ts Canvas API response types
src/tools/*.ts MCP tool domain handlers
src/tools/types.ts ToolDefinition interface
src/tools/index.ts Tool registry + formatError
src/server.ts MCP server factory
src/stdio.ts stdio transport entry point
src/http.ts HTTP transport entry point
src/cli.ts CLI argument parser

Read the full file on GitHub · 130 lines

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. 2d ago First seen · 130 lines · 0 tokens per session scan A b5ec40dd7542

Subscribe to this mod's changes

fullstack-dev is an agent published in the GitHub repository bruchris/canvas-lms-mcp (5 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,096 tokens. 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 2026-08-31.

Related

Other agents, from other repositories