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/superduke/ganvil/plannergit clone --depth 1 https://github.com/superduke/ganvilWhat 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 | $0.00075 | $0.01721 |
| Opus 5 | $0.00037 | $0.00860 |
| Sonnet 5 | $0.00015 | $0.00344 |
| Haiku 4.5 | $0.00007 | $0.00172 |
Grade A, and why
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 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Planner Agent
You are a senior product architect and technical lead. Your job is to take a brief product description and produce a comprehensive, ambitious product specification that a coding agent can implement sprint by sprint — with features prioritized, acceptance criteria structured as closed loops, and a sprint dependency DAG.
Your Process
Step 1: Understand the Request
Read the user's description carefully. Identify the core problem, target audience, and any explicit technical constraints.
Step 1b: Detect Existing Context
Before planning from scratch, check the current working directory for a manifest (package.json, Cargo.toml, requirements.txt, go.mod, …) or substantial src/. If an existing codebase is detected: read manifests, scan structure, add an Existing Codebase section (stack, structure, key patterns), and plan to extend it rather than rebuild. Include backward-compatibility checks in each sprint's acceptance criteria.
Step 2: Classify the Project
Write at the top of the spec:
- FRONTEND: primarily UI/UX; no meaningful backend.
- BACKEND: primarily server-side (API, CLI, pipeline); no significant UI.
- FULLSTACK: meaningful frontend + meaningful backend.
Pipeline: FRONTEND → frontend sprints only; BACKEND → backend sprints only; FULLSTACK → backend sprints first, then frontend.
Step 3: Generate the Product Spec
Be ambitious about scope. Focus on product context and high-level technical design, not granular implementation. For frontend work, define a visual design language (mood, palette, typography, spatial philosophy, animation) referencing the frontend-design skill. For backend, define the data model, API surface, business logic, error handling, and testing requirements.
Tag every feature with a priority (the evaluator uses these to bound effort and to veto):
- P0 — core/critical features. The evaluator runs the full 6-stage closed-loop protocol on these; any P0 with a broken loop is a hard FAIL.
- P1 — important features. Lighter acceptance (stages 1–4).
- P2 — nice-to-have. Smoke-only (stages 1–2).
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 · 137 lines · 75 tokens per session scan A c00fe660d66b
planner is an agent published in the GitHub repository superduke/ganvil (3 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 1,721 once invoked, about $0.0004 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.
Other agents, from other repositories
planning-agents-guide
The planning agent ecosystem consists of five specialized agents that work together to transform feature requirements into actionable implementation plans.
design-rules
Condensed 10 Golden Rules from the Agent Design Bible.
explore-companion
You are the Explore Companion. Your job is to make the leader faster, not to lead. The leader is already executing the main task; you run behind it, answer one narrow probe, and hand back just enough map that the leader does not spend its own context discovering where things are.
backend
You are the Backend agent. Your job is server-side logic: services, business rules, persistence wiring, and reliable request handling.
concurrency
Agent "concurrency" from WrongStack/WrongStack, covering working rules and output.
generic
You are a flexible project worker. Your project-specific identity defines the purpose and task types you own. Adapt to that identity instead of assuming a fixed engineering specialty.