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 skills add grburgess/mindgap --skill idea-courtgit clone --depth 1 https://github.com/grburgess/mindgapWrote 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/grburgess/mindgap/idea-court)<a href="https://agentmods.dev/skills/grburgess/mindgap/idea-court"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/idea-court/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.
<a href="https://agentmods.dev/skills/grburgess/mindgap/idea-court"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/idea-court.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00077 | $0.01524 |
| Opus 5.5 | $0.00031 | $0.00610 |
| Sonnet 5.5 | $0.00015 | $0.00305 |
| Haiku 4.5 | $0.00008 | $0.00152 |
Grade A, and why
idea-court 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 23d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
idea-court
Adjudicate synthesized research ideas with multi-agent fan-outs, then stage the survivors as work. Battle-tested pipeline (proven on a real research program, 2026-08): ideate → verify → gate-measure → debate → stage. A capable agent invents a plausible proposer/verifier design unaided — this skill exists to pin the calibrated shape, the grading rules that keep survivors alive, the runtime gotchas, and the graph write-back that unaided designs get wrong.
When NOT to use
Planning a survey (use deep-research); executing one (loop-system); capturing one paper (paper-to-mindmap).
The five stages
| Stage | Engine | Proven shape | Reference template |
|---|---|---|---|
| 1 Ideate | inline | mine_enrich + mine_connect on the new cluster; collide steals × gaps × in-house assets into idea-* nodes |
— |
| 2 Verify | Workflow | 5 ideas ≈ 16 agents: proposer per idea → 2 lenses per idea → 1 synthesis | references/wf-verify-ideas.js |
| 3 Gate | inline/bq | find the SHARED unmeasured assumption across survivors; run the one cheapest query that gates several ideas at once — same session, not a ticket | — |
| 4 Debate | Workflow | 4 advocates → cross-exam → claim auditor → judge ≈ 10 agents | references/wf-build-debate.js |
| 5 Stage | ingest | verdicts → node appendices; ruling → decision-*; next steps → dependency-edged proj-* with gates + first-actions |
references/graph-conventions.md |
Precedent-hunting fan-out (does an integrated system already solve the whole spec?): references/wf-precedent-fanout.js.
Live view (default-on; skip on "no live view"). Before stage 2, invoke the live-view skill: run start, give the user the URL, arm the inbox watcher (run inbox --once in the background), and install ../live-view/references/panels/verdict-matrix.js (labels must stay <lens>:<claimId>). Emit claim per idea/claim, phase.start per stage, run bind right after each Workflow launch, then reconciled verdicts, ruling, decision (go/no-go/hold), and notes for kill fork + dissent. A page flag.contest becomes a stage-2 check next round — it never flips a verdict by itself. Stage 5 ends with run report; link report.html from the decision-* node's urls.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 23d ago Changed · -1 lines e79241cc9ed8
- 27d ago Changed · +4 lines 8a7b0360b512
- 2mo ago First seen · 48 lines · 77 tokens per session scan A 3525bef1c356
idea-court is a skill published in the GitHub repository grburgess/mindgap (0 stars, last pushed 23d ago), licensed MIT. It adds 77 tokens to every session and 1,524 once invoked, about $0.0003 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 2026-09-04.
idea-court compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| idea-courtgrburgess/mindgap | Skill | 0 | 23d ago | MIT | A | 77 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |