Borrowing it
Nothing to install: this file belongs to fbabelle/PrettySeriousResearcher. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/fbabelle/PrettySeriousResearcher/main/.claude/skills/research-references/SKILL.mdgit clone --depth 1 https://github.com/fbabelle/PrettySeriousResearcherWrote 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/fbabelle/prettyseriousresearcher/research-references)<a href="https://agentmods.dev/skills/fbabelle/prettyseriousresearcher/research-references"><img src="https://agentmods.dev/badge/skills/fbabelle/prettyseriousresearcher/research-references/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/fbabelle/prettyseriousresearcher/research-references"><img src="https://agentmods.dev/badge/skills/fbabelle/prettyseriousresearcher/research-references.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 | $0.00049 | $0.01593 |
| Opus 5 | $0.00024 | $0.00796 |
| Sonnet 5 | $0.00010 | $0.00319 |
| Haiku 4.5 | $0.00005 | $0.00159 |
Grade A, and why
research-references 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 3d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research references — anti-hallucination citation gate
LLMs invent plausible-looking citations: real-sounding titles, real authors, wrong or nonexistent papers. This skill makes that impossible to ship by forcing every reference through verification before it can be cited. The rule is simple: a citation that cannot be resolved to a real source whose content supports the claim does not go in the paper.
Run this whenever citations are added or a related-work / bibliography section is built or reviewed (Phase 1 prior-art, Phase 4 writing). Use the host agent's current web-search and page-fetch tools for discovery; this skill owns verification and the verdict.
The per-reference protocol (apply to EVERY citation)
For each proposed reference — and each in-text claim that needs one — work through these in order. Do not skip ahead; an early failure is itself the verdict.
- Resolve a real link. Search for a resolvable identifier — DOI, arXiv ID, ACL Anthology / OpenReview / Semantic Scholar / publisher URL, or SSRN/NBER for finance. Investigate until you find a link or have genuinely exhausted reasonable searches. Do not stop at "this sounds like a real paper."
- Confirm it exists as cited. Title, authors, year, and venue must match the citation. A near-match (right authors, wrong year; right title, wrong venue) is a defect to correct, not a pass — fix the metadata to the resolved source. Author-collision check: when a surfaced work's author list matches one of the project's own authors, confirm with the user whether it is theirs before using it as independent evidence — self-citations are fine but must be labelled as such (and anonymized per venue rules).
- Verify content. Fetch the abstract (and the relevant section if the claim is specific). Confirm the source actually supports the specific claim it's attached to. Topical overlap is not support — "uses transformers" ≠ "shows transformers beat LSTMs on this task." If the source doesn't say what the citation implies, the citation is wrong even if the paper is real.
- Verify relevance. The work must be relevant to this paper's specific claim/context, not merely to the broad field. A correct-but-irrelevant citation is padding — drop it.
- Check coverage, currency, and status. Do not exclude a relevant work because of age. Cover the canonical origin when it matters, the closest prior work, and current representative work or a recent synthesis. Check the version of record plus corrections, expressions of concern, retractions, and material superseding results. See references/recency-and-seminal.md.
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.
- 3d ago Changed 07e8a73c2b26
- 8d ago First seen · 67 lines · 49 tokens per session scan A 1f9fbe285cd8
research-references is a skill published in the GitHub repository fbabelle/PrettySeriousResearcher (2 stars, last pushed today), licensed Apache-2.0. It adds 49 tokens to every session and 1,593 once invoked, about $0.0002 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…