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 papers-librarygit 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/papers-library)<a href="https://agentmods.dev/skills/grburgess/mindgap/papers-library"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/papers-library/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/papers-library"><img src="https://agentmods.dev/badge/skills/grburgess/mindgap/papers-library.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 | $0.00117 | $0.01341 |
| Opus 5.5 | $0.00047 | $0.00536 |
| Sonnet 5.5 | $0.00023 | $0.00268 |
| Haiku 4.5 | $0.00012 | $0.00134 |
Grade A, and why
papers-library 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
papers-library
Turn a Papers (papersapp.com / ReadCube) library into living graph structure: import →
link → discover → ideate. Self-contained: a stdlib BibTeX/RIS parser ships with this skill;
the only graph dependency is the bundled paper-to-mindmap ingest protocol.
1 · Get the library out of Papers
There is no public Papers API, so work from an export (BibTeX or RIS — both supported):
- In Papers: Settings → Export → BibTeX/RIS for the whole library, OR the gear next to All Papers → Export as → BibTeX (.bib); a collection can be exported the same way.
- Or the auto-generated
.bib: the Papers desktop app keeps a.bibper library inside the Papers Library folder (default~/Documents/Papers Library/). Point the skill at it. - Legacy Papers 3 (macOS): export to
.bib/.risfirst (File → Export), or script it via AppleScript. The local SQLite DB (Database.papersdb) is a brittle last resort — prefer an export.
Ask the user for the path to the .bib/.ris file if they did not provide one. Exports carry
references only, not PDFs — that is fine; this skill works from metadata.
2 · Pipeline (P1–P4)
P1 — Parse
python3 <skill>/scripts/parse_refs.py <export.bib|.ris> → a JSON array of normalized records
(title, authors[], year, doi, arxiv, url, abstract, keywords[], type). Auto-detects format.
Report the count (and any dropped-no-id count). For a large library, --limit to sample first.
P2 — Ingest (dedup-first)
Snapshot the graph once at the start (mindgap export --out artifacts/pre.json). Then, per record:
- Dedup BEFORE writing —
mindgap find "<arxiv id|doi>"thenmindgap find "<title>"(single salient terms are most reliable). Skip any paper already present (match on arXiv id, DOI, or normalized title); never double-ingest. - Ingest new papers via the bundled
paper-to-mindmapprotocol: one nodetype: paper, stable kebab-caseid(prefer the arXiv id or a title slug), body leading with**Authors:** …then a 2–4 sentence distillation of the abstract (what it does + why it matters);tagsfrom the record's keywords plus apapers-librarytag;urls= the arXiv, DOI, and/or source URL from the record (kindsarxiv/web).created_by: "skill:papers-library". - Link within the import + to the graph: wiki-link /
edges(relrelates_to,cites) between library papers that share a method or cite each other, and to ≥1 pre-existing node where a real relation holds. Only use URLs/relations grounded in the record or a resolved source — never a fabricated link. - Batch large libraries: ingest in chunks (e.g. 25), deduping each chunk against the live graph, so later chunks link to earlier ones.
What ships with it
6 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 dd577501a963
- 2mo ago Changed · -9 lines 3a0352bd1021
- 2mo ago First seen · 88 lines · 117 tokens per session scan A 554536830466
papers-library is a skill published in the GitHub repository grburgess/mindgap (0 stars, last pushed 23d ago), licensed MIT. It adds 117 tokens to every session and 1,341 once invoked, about $0.0005 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 2026-08-31.
papers-library compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| papers-librarygrburgess/mindgap | Skill | 0 | 23d ago | MIT | A | 117 | 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 |