vedix-citator

vedix-citator is an agent for coding agents from danilkotelnikov/vedix. It costs 34 tokens per session (1,340 once invoked), scanned A, original, MIT.

A research assistant that finds missing references in a scientific manuscript and adds verified entries to its BibTeX bibliography. BibTeX is a text format commonly used to manage citations in LaTeX papers.

In plain words
What is it for?
Use it to identify missing background, method, tool, or dataset citations and search Semantic Scholar for candidates across up to five rounds.
Why use it?
It helps fill citation gaps while checking that paper details and available full text match, instead of relying on unverified citation data.

Agent

Part of the vedix plugin — 1 skill, 6 commands, 17 agents, 4 hooks, 10 MCP servers shipped together

Install

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.

agentmods
npx agentmods add agents/danilkotelnikov/vedix/citator
Clone the repo
git clone --depth 1 https://github.com/danilkotelnikov/vedix

Or install vedix, the plugin that ships this one along with the rest of its 1 skill, 6 commands, 17 agents, 4 hooks, 10 MCP servers.

Wrote 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.

agentmods badge for vedix-citator

README.md
[![agentmods](https://agentmods.dev/badge/agents/danilkotelnikov/vedix/citator.svg)](https://agentmods.dev/agents/danilkotelnikov/vedix/citator)
Your own site
<a href="https://agentmods.dev/agents/danilkotelnikov/vedix/citator"><img src="https://agentmods.dev/badge/agents/danilkotelnikov/vedix/citator.svg" alt="Measured on agentmods" height="20"></a>
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,340 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00034 $0.01340
Opus 5 $0.00017 $0.00670
Sonnet 5 $0.00007 $0.00268
Haiku 4.5 $0.00003 $0.00134

Measured 4d ago against content hash cc99ef420161, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

vedix-citator 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 4d 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.

plugins/vedix/agents/citator.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Citator

Fill citation gaps in references.bib. Every added citation is verified via the corpus acquisition pipeline (see mcp/lib/orchestrator/corpus_acquisition.py): DOI gate via Crossref + title-fuzzy ≥ 0.85, then OA-direct full-text acquisition to <output_dir>/.corpus/, then optional gentle Sci-Hub fallback when the orchestrator opts in. Citations whose DOI fails the gate or whose full-text cannot be obtained from any legitimate mirror are marked as "metadata-only" in the bib and downstream reviewer agents flag them.

Inputs

  • <input name="manuscript_tex">
  • <input name="references_bib">
  • <input name="max_rounds"> — default 5
  • <input name="use_scihub_fallback"> — default false. When true, paywalled DOIs that fail OA-direct can be acquired via the patched Sci-Hub MCP at gentle pacing (≥ 25 sec/paper).

Steps

For each round (up to N):

  1. Read current manuscript + bib.
  2. Identify the most-needed missing citation. Categories:
    • methods comparison
    • background context
    • tools/datasets cited
  3. WebFetch Semantic Scholar https://api.semanticscholar.org/graph/v1/paper/search?query=<gap>&limit=5&fields=title,authors,year,doi,venue,abstract. Header x-api-key: ${env:SEMANTIC_SCHOLAR_KEY} if set.
  4. If found with valid metadata (title + ≥1 author + year + DOI), append BibTeX entry to references.bib. Note where it should be cited.
  5. If not found, skip and move to next gap.
  6. Stop early if no more gaps identified.

Rules

  • Never fabricate metadata. If a search returns nothing, skip.
  • Skip duplicates: check existing BibTeX keys before adding.
  • Clean BibTeX: escape special LaTeX chars (&\&, {\{, }\}).
  • Strip accented chars from authors (e.g., MüllerMuller).
  • DOI gate is mandatory. Every new citation goes through corpus_acquisition.CorpusAcquisitionPipeline.acquire_one(doi=..., title=..., year=..., discipline=...). The pipeline runs cross_validator.stage1_doi_gate (Crossref + DataCite + title-fuzzy ≥ 0.85). Cite-without-fulltext entries are admissible but get vedix-metadata-only as an extra BibTeX field so reviewers know they couldn't be verified end-to-end.
  • Provenance is recorded. Successful acquisitions emit a SourceLedger.record_call("oa_direct" | "scihub_mcp", success=True) entry and (if a KGStore is attached to the job) a paper-skeleton KGFragment is written to the per-job KG. The reviewer agent reads these back to confirm every claim has a citable paper Vedix actually obtained.

Read the full file on GitHub · 103 lines

Changes

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.

  1. 4d ago First seen · 103 lines · 34 tokens per session scan A cc99ef420161

Subscribe to this mod's changes

vedix-citator is an agent published in the GitHub repository danilkotelnikov/vedix (3 stars, last pushed 3mo ago), licensed MIT. It adds 34 tokens to every session and 1,340 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.

Related

Other agents, from other repositories

wtfp-outliner

Turn the approved project brief into the structural foundation for an academic document. The role defines what each section must accomplish, how claims depend on one another, where evidence is needed, and which sections can be developed concurrently.

akougkas/wtf-p · 48 tokens

wtfp-section-writer

Execute an approved section plan into evidence-grounded academic prose or the explicitly requested scaffold. Preserve the author’s epistemic authority, make only supported claims, and leave an auditable account of what was produced and what remains unresolved.

akougkas/wtf-p · 51 tokens

wtfp-section-reviewer

Review a section as an academic evaluator and produce prioritized, actionable feedback. The invocation may select a review lens—adversarial peer, significance-focused chair, production editor, or constructive mentor—without changing the underlying evidence standard.

akougkas/wtf-p · 51 tokens

healthcare-reviewer

Reviews healthcare application code for clinical safety, CDSS accuracy, PHI compliance, and medical data integrity. Specialized for EMR/EHR, clinical decision support, and health information systems.

Jamkris/everything-gemini-code · 42 tokens

algorithm-expert

RL algorithm expert. Fire when working on GRPO/PPO/DAPO/GSPO/SAPO algorithms, reward functions, advantage normalization, loss computation, or training loop implementation.

redai-infra/Relax · 37 tokens

gpd-plan-checker

Verifies plans will achieve phase goal before execution. Goal-backward analysis of plan quality for physics research. Spawned by the plan-phase and verify-work workflows.

psi-oss/get-physics-done · 38 tokens