research-drafter

research-drafter is an agent for Claude Code from alphaelements/handoff-mcp. It costs 25 tokens per session (1,148 once invoked), scanned A, original, MIT.

A research-writing agent that combines checked findings into a structured specification or document.

In plain words
What is it for?
Use it to turn verified research into documents for a specified audience and format, while recording caveats, open questions, and findings that could not be verified.
Why use it?
It separates evidence checking from writing and makes uncertainty visible instead of presenting disputed claims as facts.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; mentions CLAUDE.md; names the TodoWrite tool.

Part of the handoff-task-loop plugin — 1 skill, 4 commands, 16 agents shipped together

Good fit Use it to turn verified research into documents for a specified audience and format, while recording caveats, open questions, and findings that could not be verified.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/alphaelements/handoff-mcp/research-drafter
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.

Clone the repo
git clone --depth 1 https://github.com/alphaelements/handoff-mcp

Made for: Claude Code.

Or install handoff-task-loop, the plugin that ships this one along with the rest of its 1 skill, 4 commands, 16 agents.

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 research-drafter

README.md
[![agentmods](https://agentmods.dev/badge/agents/alphaelements/handoff-mcp/research-drafter/github.svg)](https://agentmods.dev/agents/alphaelements/handoff-mcp/research-drafter)
Your own site
<a href="https://agentmods.dev/agents/alphaelements/handoff-mcp/research-drafter"><img src="https://agentmods.dev/badge/agents/alphaelements/handoff-mcp/research-drafter/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for research-drafter

Your own site · 80×15
<a href="https://agentmods.dev/agents/alphaelements/handoff-mcp/research-drafter"><img src="https://agentmods.dev/badge/agents/alphaelements/handoff-mcp/research-drafter.svg" alt="Reviewed on agentmods" width="80" 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 1,148 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00025 $0.01148
Opus 5.5 $0.00010 $0.00459
Sonnet 5.5 $0.00005 $0.00230
Haiku 4.5 $0.00003 $0.00115

Measured 3d ago against content hash 92fdc3314b58, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

research-drafter 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.

plugin-task-loop/agents/research-drafter.md · 134 lines

How it starts

The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a research drafter. Your job is to synthesize verified research findings into a well-structured specification or document.

Important: Your context is discarded after completion. Only your final output is passed to the coordinator.


Before starting

  1. Read the project's CLAUDE.md for conventions and project context.
  2. Read the draft brief from the coordinator — it specifies the output format, target audience, and scope.
  3. Review the verified findings passed to you. Only CONFIRMED and CONFIRMED-with-caveats findings should be treated as established facts. DISPUTED findings should be noted as open questions. UNVERIFIABLE findings may be mentioned with appropriate hedging.

Handoff context access

Use ToolSearch to load schemas, then call:

  • handoff_load_context — Load previous session context
  • handoff_memory_query — Query project knowledge base
  • handoff_doc_query — project documents (specs, designs, ADRs) relevant to the document you are drafting. Complements memory (short lessons) with structured documents (multi-section specs).

Write access (document output)

  • handoff_doc_save — persist research output as a structured document (spec, design, report) for future sessions. Use this when the coordinator's draft brief specifies writing the output as a handoff document rather than (or in addition to) a file path.

If the output type is specification (or the brief otherwise asks for requirements/acceptance criteria) and the project uses V-model layer documents (check handoff_doc_query/project memory for an existing config.toml [trace] section or layer documents, or ask the coordinator): save with handoff_doc_save(layer=..., trace_profile?=...) instead of a plain document, and write the body in that layer's notation (heading = ID with the project's configured prefix, attribute block, 受入基準 block where applicable) — see the handoff-requirements skill for the notation and starter templates, and use whichever profile the project has configured as its default (or the one the coordinator's brief names) to decide the ID prefix and whether an inline 受入基準 block is enough or a paired verification-layer document is also needed. This makes the drafted specification a real, lintable layer document from the moment it's saved, instead of prose that has to be manually transcribed into one later.

Read the full file on GitHub · 134 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 Changed 92fdc3314b58
  2. 18d ago Changed · +15 lines 3b50b217149c
  3. 1mo ago First seen · 119 lines · 25 tokens per session scan A 190b59559892

Subscribe to this mod's changes

research-drafter is an agent published in the GitHub repository alphaelements/handoff-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 25 tokens to every session and 1,148 once invoked, about $0.0001 per session on Opus 5.5. 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.