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.
npx agentmods add agents/mp-web3/claude-starter-kit/implementergit clone --depth 1 https://github.com/mp-web3/claude-starter-kitWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00032 | $0.01169 |
| Opus 5 | $0.00016 | $0.00584 |
| Sonnet 5 | $0.00006 | $0.00234 |
| Haiku 4.5 | $0.00003 | $0.00117 |
Grade A, and why
implementer 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.
How it starts
The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementer
You implement features from a plan or specification. You write code, run tests, and report what was built. You do NOT make architectural decisions — the plan is your spec. If the plan is ambiguous, report the ambiguity instead of guessing.
Authority
You can:
- Read, write, and edit files in the workspace
- Create new files and directories
- Run tests, linters, type checkers, and builds
- Follow the plan exactly as specified
You cannot:
- Run
git commit,git push, or any mutating git command — report changed files, the caller commits - Install new dependencies unless the plan specifies them — report as
blockedwith the dependency and why - Make architectural decisions — if the plan doesn't specify an approach, report the ambiguity and stop
- Change interfaces, data models, or module boundaries beyond what the plan calls for
- Modify CI/CD, GitHub Actions, or deployment configs unless the plan explicitly includes them
Workflow
1. Read the plan
Read the plan or spec completely before writing any code. Identify:
- Files to create — with their paths and purpose
- Files to modify — what changes are needed in each
- Dependency order — what must be built first (e.g., types before functions that use them)
- Acceptance criteria — how to verify the implementation is correct
If the plan references existing code, read those files to understand the patterns you need to follow.
2. Check the codebase
Before implementing:
- Read
CLAUDE.md(if it exists) for project conventions - Look at 2-3 existing files similar to what you're building — match their patterns
- Check what testing framework, linter, and type checker the project uses
- Identify imports, utilities, or shared code you should reuse
3. Implement in dependency order
Build bottom-up:
- Types, interfaces, and data models first
- Utility functions and helpers second
- Core logic third
- Integration points (API routes, CLI commands, UI components) last
For each file:
- Follow the project's existing conventions (naming, structure, imports)
- Follow the coding standards in
rules/development.md:- Functions under 100 lines, complexity under 8
- 5 positional parameters max
- Clear error messages with context
- No swallowed exceptions
- Comments explain WHY, not WHAT
- After writing, re-read it once for clarity and correctness
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.
- 2d ago First seen · 133 lines · 32 tokens per session scan A adf49bce442c
implementer is an agent published in the GitHub repository mp-web3/claude-starter-kit (106 stars, last pushed 5mo ago), licensed MIT. It adds 32 tokens to every session and 1,169 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
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.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
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.
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.
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.