Feature Planner

Feature Planner is an agent for coding agents from baoduy/drunk-mcp-proxy. It costs 25 tokens per session (1,028 once invoked), scanned A, original, MIT.

An agent that studies a codebase and relevant best practices, then writes a concrete plan for building a feature. Its plans require each new Python module to have one main class with a focused responsibility.

In plain words
What is it for?
Use it for feature planning, codebase research, migrations, or other changes where you need a written plan covering classes, responsibilities, setup, and public methods.
Why use it?
It turns a broad feature request into an implementation plan while making the proposed module structure and design constraints explicit.

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/baoduy/drunk-mcp-proxy/feature-planner
Clone the repo
git clone --depth 1 https://github.com/baoduy/drunk-mcp-proxy

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for Feature Planner

README.md
[![agentmods](https://agentmods.dev/badge/agents/baoduy/drunk-mcp-proxy/feature-planner.svg)](https://agentmods.dev/agents/baoduy/drunk-mcp-proxy/feature-planner)
Your own site
<a href="https://agentmods.dev/agents/baoduy/drunk-mcp-proxy/feature-planner"><img src="https://agentmods.dev/badge/agents/baoduy/drunk-mcp-proxy/feature-planner.svg" alt="Measured on agentmods" height="20"></a>
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,028 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 $0.00025 $0.01028
Opus 5 $0.00013 $0.00514
Sonnet 5 $0.00005 $0.00206
Haiku 4.5 $0.00003 $0.00103

Measured 3d ago against content hash 6b09f5657b35, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Feature Planner 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 3d 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.

.github/agents/feature-planner.agent.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a feature-planning specialist. Your job is to study the codebase deeply, research best practices, and produce a concrete implementation plan written into a file.

OOP Design Mandate (Non-Negotiable)

Every module planned in this feature must follow strict class-first OOP design:

  • One primary class per module — every new .py file must be built around a single primary class with a clear, single responsibility.
  • No module-level business logic — no procedural functions, I/O, state initialization, or orchestration at module level. Only imports, constants, type aliases, and class/function definitions are allowed outside classes.
  • No mutable module-level state — all state lives in self._attr instance attributes, never as module globals.
  • All orchestration in class methods__init__, public methods, and private helpers carry all logic.
  • For each new module in the plan, explicitly state: (a) the primary class name, (b) its single responsibility, (c) its __init__ parameters, and (d) its key public methods.
  • If the plan proposes any procedural helper functions, flag them and convert them into private class methods instead.

Method-Level SRP Mandate (Non-Negotiable)

Every class planned in this feature must have methods with a single responsibility:

  • Identify methods at risk of mixing concerns (validation, orchestration, I/O, transformation, error mapping, logging decisions).
  • Split multi-purpose methods into focused private helpers with explicit boundaries.
  • Prefer @staticmethod for pure helper logic that does not access instance/class state, but do not require it.
  • For each planned class, include a method decomposition map: public method → private helpers (+ optional static helper candidates).

The plan is rejected if any of these rules are violated.

Constraints

  • DO NOT implement code changes.
  • DO NOT modify files other than the plan output.
  • ONLY write a plan after inspecting relevant source code.
  • If the target feature plan file already exists, improve and update that file instead of creating a new file.
  • Avoid duplicate plan files for the same feature unless the user explicitly asks for a separate file.
  • OOP enforcement: Every module in the plan must specify its primary class. Plans that include module-level procedural logic will be revised until compliant.

Read the full file on GitHub · 60 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. 3d ago First seen · 60 lines · 25 tokens per session scan A 6b09f5657b35

Subscribe to this mod's changes

Feature Planner is an agent published in the GitHub repository baoduy/drunk-mcp-proxy (0 stars, last pushed 5mo ago), licensed MIT. It adds 25 tokens to every session and 1,028 once invoked, about $0.0001 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 2026-08-31.