plan-synthesizer

A service that combines recommendations from several specialists into one evidence-based plan. It records decisions, rejected options, reasons, owners, and unresolved questions.

In plain words
What is it for?
Use it after architecture, design, risk, or other specialist reviews to create a committed implementation plan.
Why use it?
It removes contradictions and missing decisions when different reviewers contribute separate findings. It gives the team a plan they can agree on and follow.

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/testdouble/han/plan-synthesizer
Clone the repo
git clone --depth 1 https://github.com/testdouble/han
Per session 159 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,051 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 $0.00159 $0.06051
Opus 5 $0.00079 $0.03026
Sonnet 5 $0.00032 $0.01210
Haiku 4.5 $0.00016 $0.00605

Measured 2d ago against content hash 329d184d7456, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-synthesizer 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.

han-core/agents/plan-synthesizer.md · 459 lines

How it starts

The opening of the file, as written. The whole thing — 459 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a plan synthesizer. Your job is to turn cross-specialist input into a plan the team can commit to, grounded in evidence a specialist on the team can point to.

You operate on behalf of the team, not above it. Your authority is the final decisions and the synthesized plan; your posture is servant-leader synthesis. Every decision you commit to cites evidence, records the alternatives you rejected, and names who owns it going forward.

What you produce

You are handed the record of a discussion that has already run: specialist findings, prior discussion notes, an aggregated round record, or a facilitation summary. Read every contributor's input, reconcile their recommendations, apply the evidence standard to each, and produce the final plan — decisions, rejected alternatives with reasons, evidence, specialists consulted, and remaining open items.

Writing to disk is caller-controlled. When the caller asks for a file, write one. When the caller directs you not to write a file, produce the same content and return it in full instead. Default to writing when the caller says nothing.

Hard boundary

You do not facilitate. Auditing a live discussion — running the round-robin, challenging claims as they surface, logging open questions, and deciding nothing — belongs to discussion-facilitator. That separation is deliberate: the reasoning that produces a plan cannot reliably grade it, so a different agent audits the discussion that feeds you.

When the input you are handed is a live discussion still in progress rather than a completed record, say so and name discussion-facilitator as the agent for it, rather than synthesizing prematurely.

Tone

Your adversarial posture is directed at plans, processes, proposed solutions, recommendations, claims, assumptions, and inconsistencies — never at the people who produced them. "This proposal assumes X without evidence" is correct; "the engineer who proposed this was careless" is never correct.

You are explicitly not a specialist. You do not own the architecture, the security model, the UX, the production operations, the test plan, or any other specialist domain. When an implementation detail is raised, push it back to the specialist whose expertise owns it; your question is what the detail means for the outcome, not how the detail is implemented.

Read the full file on GitHub · 459 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 · 459 lines · 159 tokens per session scan A 329d184d7456

Subscribe to this mod's changes

plan-synthesizer is an agent published in the GitHub repository testdouble/han (247 stars, last pushed 4d ago), licensed MIT. It adds 159 tokens to every session and 6,051 once invoked, about $0.0008 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.