Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add SyloRei/claude-godmode/plugin install claude-godmodeWrote 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.
[](https://agentmods.dev/skills/sylorei/claude-godmode/refine)<a href="https://agentmods.dev/skills/sylorei/claude-godmode/refine"><img src="https://agentmods.dev/badge/skills/sylorei/claude-godmode/refine.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00086 | $0.03318 |
| Opus 5 | $0.00043 | $0.01659 |
| Sonnet 5 | $0.00017 | $0.00664 |
| Haiku 4.5 | $0.00009 | $0.00332 |
Grade A, and why
refine 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 8d 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 — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Refine
The mid-mission gap-analysis step of the planning spine. Where /ideate is pre-mission and generative — exploring directions for the next mission before any charter exists — /refine is mid-mission and analytical: it works inside the active mission, reads what has already been planned, finds what's missing or underspecified, converges on one gap, and turns it into a buildable unit. It is the sibling of /ideate, one step later in the lifecycle.
/refine is strictly additive. It reads the active mission's roadmap and briefs, surfaces candidate gaps and improvements, converges on the one most worth doing next, and appends a new numbered work unit to the roadmap plus that unit's complete brief. It never edits or rewrites an existing brief or roadmap unit in place — in-place updates are delegated to /brief N. This is the spine's gap-analysis seam: /mission → /brief N → /plan N → /build N → /verify N → /ship, with /refine feeding a new appended unit back to /brief//plan whenever a gap is found.
The artifacts live in the consumer's repo, never in the plugin source, scoped to the active mission:
.planning/missions/<mission_id>/ROADMAP.md— the new unit is appended here (existing entries untouched)..planning/missions/<mission_id>/briefs/NN-name/BRIEF.md— the new unit's complete brief, in the same format/briefproduces.
<mission_id> is the active mission's NN-slug, resolved from workflow state ("$gm/godmode-state" get mission_id; see the helper-resolution note in Process), so /refine always operates on whatever mission is currently active. .planning/PROJECT.md and .planning/STATE.md stay at the .planning/ root (project-global).
Mission-scoped contract
/refine is a mission-scoped skill. Unlike /ideate, it does not run on a bare repo:
- Requires an active mission. It resolves
mission_idfrom workflow state and operates entirely within that mission. With no active mission it stops and points the user at/mission— it never scaffolds or invents a mission. - Reads the whole mission before converging. It surveys the active mission's roadmap and existing briefs to find gaps — not just the unit the user happens to be on.
- Strictly additive — never edits in place. It appends a new unit and its brief. It never rewrites, renumbers, or edits an existing roadmap entry or brief. A gap inside an already-specified unit becomes either a new appended unit or is deferred to
/brief N(see below). Allocating amission_idand resetting the work-unit counter remain/mission's job alone —/refinedoes neither.
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.
- 8d ago First seen · 213 lines · 86 tokens per session scan A 9cb41a0eb147
refine is a skill published in the GitHub repository SyloRei/claude-godmode (3 stars, last pushed 3mo ago), licensed MIT. It adds 86 tokens to every session and 3,318 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 skills, from other repositories
figma-verification
Extract Figma design specs before implementing UI components. Use depth 5, cache to disk, extract ALL properties, map to design system tokens.
interaction-design
Use when designing how a UI feels to use. Covers audience dimensions, discoverability, mode communication, keyboard-first UX, feedback timing, empty states, and information density. Includes WCAG 2.1 AA compliance audit. Companion to visual-design.
browser-testing
Adversarial browser testing. Use when testing UI behavior, verifying network requests, checking console errors, filling forms, or handling dialogs. The goal is finding bugs, not confirming happy paths.
debugging-protocol
Systematic protocol for diagnosing broken behavior and making recurrence structurally impossible. Schema mismatch causes 80% of bugs. Use when something isn't showing, isn't working, or is still broken after a fix attempt.
plan-mode
CEO and engineering review after spec-writing. Auto-selects weight class from the spec format (skip for task-mode, eng-only for routine, full CEO+eng for high-risk). Run /spec-writing first.
pr-review
Use when reviewing a PR or responding to review comments on your own PR. Detects authorship automatically and runs the correct mode. Never makes code changes when reviewing someone else's work.