claude

paper-trail

Plugin Claude Code anti-hallucination pour la recherche scientifique. Cree des SOTAs (etats de l'art) garantis sans fabrication, audite les SOTAs/articles existants (purge ou warnings), avec un moteur de validation strict (FSM 8 etats + cascade d'acquisition PDF + page 1 anti-homonymie + 19 invariants de coherence). Workflow generique multi-domaines, adapter Obsidian par defaut. Sources d'acquisition etendues optionnelles, desactivees par defaut (opt-in explicite requis ; voir DISCLAIMER).

7 on GitHubby Romain Peyrichouv0.4.0MITupdated 2 months ago

Not in a marketplace yet, so it loads for one session with --plugin-dir.

Plugin paper-trail in roomi-fields/paper-trail. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (6)

  • citation-receipts

    Systematic per-citation verification protocol for academic writing. MUST be used whenever writing content that cites sources — papers, blog posts, SOTAs, or any document making factual claims about external work. Triggers (FR + EN) on any writing task involving references, citations, or bibliographi

  • paper-writer

    Write academic papers (IMRaD structure) for conferences and journals, with mandatory anti-hallucination citation verification. Triggers (FR + EN) : "écris un paper", "article pour conférence", "rédige pour la conférence", "structure IMRaD", "soumission paper", "write a paper on", "academic article",

  • pdf-cascade

    Acquire PDFs for bibliographic references via a strict 8-source cascade (Crossref OA → arXiv → OpenAlex → Unpaywall → HAL → CORE → archive.org → WebSearch queue; optionally Sci-Hub + Anna's Archive, plus Anna's via a windowed browser, in opt-in mode). Each acquired PDF is validated against expected

  • registry-doctor

    Audit the bibliography registry for consistency via 19 mechanical invariants (I1-I19) covering state validity, slug uniqueness, UID format, PDF path normalization, file presence on disk, sha256 integrity, page-1 validation log coherence, state history monotonicity, attempt numbering, blocked-reason

  • sota-auditor

    Final authority on the truthfulness of references cited in State-of-the-Art reviews and academic papers. Activate this skill whenever the session involves auditing, correcting, or cleaning citations in a SOTA or article, or when responding to verdicts from `pdf-cascade` / `registry-doctor`. Triggers

  • sota-writer

    Write structured State-of-the-Art (SOTA) reviews / literature reviews for academic research projects with **zero hallucinated citations**. Activate this skill whenever the user wants to write, update, or expand a SOTA. Triggers (FR + EN) : "écris un SOTA", "état de l'art sur", "revue de littérature"

Slash commands (22)

  • /paper-trail:paper-trail-acquire

    Troisième passe du pipeline cible refondu. Lance la cascade PDF (10 sources : Crossref OA + arXiv + OpenAlex + Unpaywall + HAL + CORE + éditeur par DOI + archive.org + WebSearch ; jusqu'à 17 avec les voies par navigateur et les sources étendues opt-in) ciblée sur les refs d'un SOTA donné. Différent

  • /paper-trail:paper-trail-audit-all

    Batch audit of every SOTA in the vault. For each SOTA, identifies refs that are page1_validated (technical OK) but not yet sota_cited_confirmed (content audited), and launches audit-sota on those SOTAs. Returns a per-SOTA summary at the end.

  • /paper-trail:paper-trail-audit-article

    Per-citation audit of an academic article (LaTeX or Markdown). Produces RECEIPTS.md classifying each citation as VALID/ADJUST/INVALID/UNVERIFIABLE. Optional --warn inserts inline warnings in a .bak copy of the source.

  • /paper-trail:paper-trail-audit-sota

    Audit a SOTA's bibliography. Classify each cited ref by validation state. Optional --purge removes citations to retracted refs from the SOTA (with .bak backup).

  • /paper-trail:paper-trail-cascade

    Acquire PDFs via the 8-source cascade (11 with opt-in extended sources) for a single ref by slug, or a batch filtered by state. Validates page 1 anti-homonymy on each download.

  • /paper-trail:paper-trail-decide

    Inspect a single reference in detail (identity, PDF status, acquisition history with verdicts, citation context in vault, state transitions) and decide its fate. Offers actionable decisions based on the ref's current state.

  • /paper-trail:paper-trail-doctor

    Audit the registry via the 19 mechanical invariants (I1-I19). Reports ERROR/WARN/INFO violations. Optional auto-fix for safe ones (I4 path, I6 sha256, I9 attempts).

  • /paper-trail:paper-trail-identify

    Première passe du pipeline cible refondu (read-only). Produit un rapport d'identification : pour chaque mention citée dans le SOTA, indique si la fiche existe déjà au registre, son état actuel, si elle a un PDF, et l'action recommandée (réutiliser / créer / passer en revue). Ne modifie ni le SOTA ni

  • /paper-trail:paper-trail-ingest-all

    Batch INGEST across the entire vault. Scans all SOTAs, identifies those with bibliographic sections containing free-text citations, then runs /paper-trail:ingest on each one sequentially. Reports per-SOTA summary at the end. The single batch entry point to make a legacy vault fully conform to the pa

  • /paper-trail:paper-trail-ingest

    Ingest the citations of one SOTA into the registry. Extracts bibliographic sections, parses citations via the citation-parser sub-agent, resolves DOI via Crossref/S2, deduplicates against the registry, creates new candidate refs, and substitutes free-text citations with [[wikilinks]]. Required first

  • /paper-trail:paper-trail-inject-url

    Injecte une URL OA connue (HAL, dépôt uni, NIME, page perso) pour une ref dont la cascade automatique a échoué, puis relance l'acquisition + validation page 1.

  • /paper-trail:paper-trail-linkify

    Dernière passe du pipeline cible refondu. Insère les wikilinks finaux dans le SOTA (vers PDF si validé, sinon vers ancre dans une section `## Statut des sources` régénérée idempotemment en bas du fichier). Présuppose que les passes identify/purge/acquire ont été lancées.

  • /paper-trail:paper-trail-new-paper

    Start writing an academic paper (IMRaD structure) on a topic, with anti-hallucination citation verification at every step. Builds on existing audited SOTAs.

  • /paper-trail:paper-trail-new-sota

    Create a new State-of-the-Art review on a topic, guaranteed without hallucinated citations. Inverted workflow (research → acquire PDFs → read → write), refuses to write from memory.

  • /paper-trail:paper-trail-purge

    Nettoie les wikilinks invalides d'un SOTA (vers refs retracted, refs `_0000_*` orphelines, suffixes numériques `_2_3_4`, fichiers techniques `20_ATLAS/`, `30_DEV/`, `.canvas`, slugs non-bibliographiques). Opère uniquement sur le SOTA, jamais sur le registre. Backup git auto avant `--apply`.

  • /paper-trail:paper-trail-reactivate-ocr

    Re-evaluate refs in state `awaiting_rtfm_ocr` via `rtfm check`. Transitions to `page1_validated` if OCR succeeded, or `needs_reacquisition` if OCR failed.

  • /paper-trail:paper-trail-receipts

    Local-only per-citation audit (no remote API). Reads the cited PDFs from the local registry, generates RECEIPTS.md. Faster than audit-article (no Crossref / paper-search calls).

  • /paper-trail:paper-trail-registry-cleanup

    Nettoyage historique du registre des fiches bibliographiques (les ~900 fiches existantes). Cible les vieilles fiches mal nommées (`*_0000_*`, `_untitled`) ET les duplicates avec suffixes numériques (`foo_2020_bar_2`, `_2_3_4`) qui sont des artefacts de runs INGEST passés. Délègue les décisions au su

  • /paper-trail:paper-trail-resolve-textbooks

    Resolve incomplete textbook references in the registry. Reads candidates via `pipeline resolve-textbooks --list`, invokes the `textbook-resolver` sub-agent to produce a decisions JSON (merge_into / complete / blocked), and applies via `pipeline resolve-textbooks --apply-from`. Fully automated cleanu

  • /paper-trail:paper-trail-review

    Review and arbitrate problematic references (cascade-exhausted, unresolved UID, awaiting OCR). Generates an up-to-date report, walks through cases by category, and applies arbitration decisions to the registry.

  • /paper-trail:paper-trail-search

    Search the validated registry for references matching a topic (author, title, year, slug). Returns only refs in state page1_validated or sota_cited_confirmed by default — i.e. references that have been technically validated (PDF acquired, page 1 OK) or fully audited (content validated against PDF).

  • /paper-trail:paper-trail-status

    Show the registry status — count of refs per FSM state. Read-only, fast (< 5s).

Subagents (6)

  • cascade-runner

    Sub-agent that orchestrates the PDF acquisition cascade for a batch of refs. Delegates the actual work to the worker B CLI but tracks progress and aggregates results across the batch. Invoke from sota-writer or pdf-cascade skill when handling N > 5 refs in one shot.

  • citation-parser

    Sub-agent that parses bibliographic sections and inline citations from SOTA / article text. Takes raw text (a section header + content, or an inline excerpt) and returns structured JSON `[{author, year, title, doi?, venue?, raw}]`. Isolates the LLM extraction from the main agent context. Invoke from

  • claim-checker

    Sub-agent that audits a specific claim against the PDF cited. Invoked by citation-receipts skill for deep PDF↔claim verification. Returns structured verdict (VALID/ADJUST/INVALID/UNVERIFIABLE) with evidence quoted from the source. Isolates the heavy PDF reading from the main agent's context.

  • page1-validator

    Sub-agent that validates whether a downloaded PDF matches the expected metadata (author, title, year). Anti-homonymy check on page 1. Invoke when a manual page 1 verification is needed on an acquired PDF, separate from the cascade's automatic validation.

  • researcher

    Sub-agent that performs exhaustive multi-source academic search (paper-search MCP across 22 platforms + optional NotebookLM + optional WebSearch). Returns structured JSON of candidate refs for sota-writer phase A. Invoke when broad literature search is needed without polluting the main agent's conte

  • textbook-resolver

    Sub-agent that decides what to do with incomplete textbook references in the registry. Takes a JSON list of candidates (refs with year=0000 or title=empty, plus their siblings = other refs with same lastname) and returns JSON decisions (merge_into, complete, or blocked). Invoked by /paper-trail:reso

Hooks (5)

  • PreToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/tools/precheck_sota_wikilinks.py"

  • PreToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/pre_save_sota_check.py"

  • PostToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/post_edit_ref_check.py"

  • PostToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/post_edit_sota_check.py"

  • SessionEnd

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/session_end_doctor.py"

View source on GitHub

Similar plugins

humanizer

claude

Rewrite AI-sounding text so it reads like the writer without changing what it says.

1 skill

blader/humanizer 68,461

5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.

5 skills · 16 commands · 4 agents · hooks

Imbad0202/academic-research-skills 56,698

OpenAPI specification generation, Mermaid diagram creation, tutorial writing, API reference documentation

3 skills · 1 command · 5 agents

wshobson/agents 42,461

openseo

claudecodex

Without good data, your agent gives generic advice. Keyword research, competitor research, backlinks, and site audits. All for $10/month.

12 skills · 1 MCP server

every-app/open-seo 27,513

promptfoo

claudecodex

Connect providers and targets, write eval suites, and set up or run red team scans with promptfoo

4 skills

promptfoo/promptfoo 27,115

prowler

claude

Connect Claude to Prowler Cloud through the Prowler MCP server. Reads providers, scans, findings, compliance results and resources for AWS, Azure, GCP, Kubernetes, Microsoft 365, GitHub and the other providers listed at https://registry.prowler.com, and returns remediation guidance for failed checks. Writes are limited to Prowler Cloud itself: triggering scans, finding triage, mute rules and the mutelist, providers and integrations. Your cloud resources are not changed by the connector.

1 skill · 1 MCP server

prowler-cloud/prowler 15,234

Maintain paper-trail?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted