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 roomi-fields/paper-trail --skill citation-receiptsgit clone --depth 1 https://github.com/roomi-fields/paper-trailWrote 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/roomi-fields/paper-trail/citation-receipts)<a href="https://agentmods.dev/skills/roomi-fields/paper-trail/citation-receipts"><img src="https://agentmods.dev/badge/skills/roomi-fields/paper-trail/citation-receipts/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/roomi-fields/paper-trail/citation-receipts"><img src="https://agentmods.dev/badge/skills/roomi-fields/paper-trail/citation-receipts.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 | $0.00174 | $0.02648 |
| Opus 5 | $0.00087 | $0.01324 |
| Sonnet 5 | $0.00035 | $0.00530 |
| Haiku 4.5 | $0.00017 | $0.00265 |
Grade A, and why
citation-receipts 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 10d 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 — 273 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill : Citation Receipts
Why this skill exists
The plugin's mechanical safeguards (FSM, page 1 anti-homonymy, cascade) prevent citing a fabricated reference. But they don't prevent citing a real reference for a claim it doesn't actually make. Examples that have occurred :
- Inverted result : citing a paper as proving NP-hardness when it proves tractability
- Fabricated quote : putting words in quotation marks that don't appear in the cited source
- Wrong attribution : citing the wrong author for a known result (attribution chains)
- Equivalent ≠ formulated as : saying X "is equivalent to" Y is much stronger than X "is formulated as" Y
This skill enforces a per-citation read-before-cite protocol. Inspired
by the receipts plugin (MIT, James Weatherhead) for the audit-PDF↔claim
pattern.
The golden rule
Read the source before you cite it. Not the title. Not the abstract. Not what you think it probably says based on the author's reputation. The actual text where the result is stated. If you can't access the source, mark it UNVERIFIABLE explicitly — don't guess.
When this skill activates
This protocol applies to EVERY citation in :
- Research papers (
paper-writerskill output) - SOTAs / literature reviews (
sota-writerskill output) - Blog posts and other academic content
/paper-trail:audit-article <path>— runs this skill on an existing article file/paper-trail:receipts <path>— local audit (no remote API calls)- Pre-submission audit checklist
Verification workflow (per citation)
For each source you intend to cite (or that's already cited in the document under audit), follow these steps in order.
Step 1 — Locate the source
Check these locations, in order :
- Local registry :
$RESEARCH_REGISTRY_PATH/refs/<slug>.md. If state ispage1_validatedor higher, the PDF is at$RESEARCH_SOURCES_PATH/<pdf_path>. paper-searchMCP for not-yet-acquired sources :mcp__paper-search__search_papers(query=<title>)thenget_crossref_paper_by_doi(<doi>)for confirmed metadatanotebooklmMCP (optional, if configured) for book chapters- Not found anywhere → mark UNVERIFIABLE, flag to user, do not proceed with the citation as-is
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.
- 10d ago First seen · 273 lines · 174 tokens per session scan A 413d67059dfe
citation-receipts is a skill published in the GitHub repository roomi-fields/paper-trail (5 stars, last pushed 7d ago), licensed MIT. It adds 174 tokens to every session and 2,648 once invoked, about $0.0009 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
draft
Use to produce a first draft, or to revise one from an upstream verdict artifact's open-defect count, emitting a draft@1 artifact. The target node of a bounded back-edge. Single tool. Reusable standalone.
qiongli
Qiongli version: v1.17.0. Cross-platform academic research workflow for Codex, Claude / Claude Code, and CLI. Use for academic research lifecycle work: paper planning, literature review, paper reading, gap finding, study design, manuscript writing, statistics, analysis code, reproducibility, proofread, rebuttal…
x-bug2rag
A knowledge-capture tool that turns reusable bug explanations into a local RAG collection, meaning a searchable store of text that an agent can retrieve later. It records the trigger, incorrect implementation, correct implementation, and observable difference.
x-cr
A software-correctness investigation skill for finding why code behaves differently from what was expected. It uses evidence from code paths, specifications, tests, logs, and changes to assess possible causes.
x-spec2
A compact system-design guide for turning vague or cross-module requests into a small, structured specification package. It defines requirements, acceptance scenarios, module boundaries, and—when needed—data flow, state, timing, resources, or recovery design.
x-adversarial-risk
A focused adversarial review of a software specification. It tries to find small counterexamples that would expose incorrect implementations, such as invalid state changes, crashes, duplicate actions, permission mistakes, or concurrent events.