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 deep-researchgit 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/deep-research)<a href="https://agentmods.dev/skills/grburgess/mindgap/deep-research"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/deep-research/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/deep-research"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/deep-research.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 | $0.00000 | $0.00911 |
| Opus 5.5 | $0.00000 | $0.00364 |
| Sonnet 5.5 | $0.00000 | $0.00182 |
| Haiku 4.5 | $0.00000 | $0.00091 |
Grade A, and why
deep-research 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
deep-research
Deep-think research planning. Brainstorming asks good questions; second-brain knows what's already known; the mining modes know what isn't. This skill wires them: plans are grounded in graph evidence and shaped by real gaps, not naive clarifiers. It plans — it never executes the research.
Dispatch
- Inside a loop-system session → stop (loops own their own protocol).
- Build-shaped ask → hand back to superpowers:brainstorming. The reverse entry is expected: brainstorming escalates HERE when an ask proves research-shaped ("we don't know enough to design this").
- Otherwise run the phases: GROUND → GAP → DIVERGE → [DEEP THINK] →
CONVERGE → PAY BACK. Full protocol:
references/phases.md.
Live view (default-on; skip on "no live view"). At GROUND invoke mindgap:live-view
(run start --skill deep-research), give the user the URL, arm the inbox watcher (run inbox --once in the background). Emit
phase.start per phase, claim per gap (G<n>) — every DIVERGE question and its answer is
mirrored by the askuser hook (the page shows how long each waited on the user) — run bind for a DEEP THINK Workflow, and at CONVERGE a verdict per gap —
ADDRESSED (text names the RQ), COVERED (by evidence), DEFERRED, or STRUCK — plus
artifact for the plan doc. PAY BACK: decision go (approved) / hold (revise), run.end,
run report, and add report.html to the research-plan node's urls. A page
flag.contest on a gap reopens it as a DIVERGE question.
The phases, one line each
- GROUND — second-brain RETRIEVE on the topic (enrich subgraph is the centerpiece).
- GAP — mine
learn+ loop open-failures + contradictions → an evidence-cited gap list, shown to the user as the agenda. - DIVERGE — superpowers:brainstorming dialogue; every question cites evidence or a gap item.
- DEEP THINK — offer only, when directions are contested or evidence thin: read-only scout fan-out + judge panel. Never auto-fired.
- CONVERGE —
<project>/docs/research/YYYY-MM-DD-<topic>.md: questions, hypotheses, method, evidence, measurable done-criteria, loop-shaped verdict. Approval-gated. - PAY BACK — register the plan in the graph (always), distill learnings, and if loop-shaped, offer loop-system INIT with §1–§2 pasted in.
What ships with it
1 file 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 d0e10f1c3d81
- 27d ago Changed · +11 lines 6a5e3e233913
- 2mo ago First seen · 47 lines · 0 tokens per session scan A 48f585677aff
deep-research is a skill published in the GitHub repository grburgess/mindgap (0 stars, last pushed 23d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 911 tokens. 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.
deep-research compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| deep-researchgrburgess/mindgap | Skill | 0 | 23d ago | MIT | A | 0 | 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 |