{
  "markdown": "# ARS-Codex\n\n[![Version](https://img.shields.io/badge/version-v0.1.28-blue)](VERSION)\n[![License: CC BY-NC 4.0](https://img.shields.io/badge/license-CC%20BY--NC%204.0-lightgrey)](https://creativecommons.org/licenses/by-nc/4.0/)\n[![Sponsor](https://img.shields.io/badge/sponsor-Buy%20Me%20a%20Coffee-orange?logo=buy-me-a-coffee)](https://buymeacoffee.com/crucify020v)\n\n**Languages:** [English](README.md) · [简体中文](README_ZH-CN.md) · [繁體中文](README_ZH-TW.md) · [日本語](README_JA.md)\n\nARS-Codex is the Codex-native sibling of\n[Academic Research Skills (ARS) for Claude Code](https://github.com/Imbad0202/academic-research-skills).\nIt is a separate Codex distribution with its own plugin identity, packaging,\nversioning, and runtime adapter.\n\nThis repository vendors the ARS workflow content as a single Codex skill:\n\n```text\nskills/academic-research-suite/\n  SKILL.md\n  manifest.json\n  agents/openai.yaml\n  codex/\n    full-runtime-manifest.json\n    agents/\n    hooks/\n    scripts/\n  ars/\n    deep-research/\n    academic-paper/\n    academic-paper-reviewer/\n    academic-pipeline/\n    experiment-agent/\n    commands/\n    hooks/\n    docs/\n    tests/\n    shared/\n```\n\nThe original Claude Code ARS checkout is not modified. Upstream content is copied\nfrom fresh GitHub clones and adapted through the Codex router in\n`skills/academic-research-suite/SKILL.md`.\n\n## Relationship to Claude Code ARS\n\nThis repository is ARS-Codex. For the original Claude Code ARS distribution, use\n[Imbad0202/academic-research-skills](https://github.com/Imbad0202/academic-research-skills).\n\nUse the Claude Code repo when you want the native Claude Code skill layout,\nClaude-specific agent-team behavior, or the original ARS development history.\nUse this repo when you want the Codex-native single-suite skill.\n\n## Versioning\n\nThis ARS-Codex package is version `0.1.28`. The repo-root `VERSION` file,\n`skills/academic-research-suite/SKILL.md` metadata version, and\n`skills/academic-research-suite/manifest.json` `adapter_version` track the\nCodex package version independently of the vendored ARS suite. Vendored upstream\nversions are recorded by commit in `manifest.source_repositories[]`.\n\nPackage-level changes are summarized in [`CHANGELOG.md`](CHANGELOG.md).\n\nThe vendored ARS source currently tracks upstream `main` after the signed\n`v3.21.1` release at\n`Imbad0202/academic-research-skills@94436237913091d4739870159d241660527e8338`\n(2026-09-02; upstream suite metadata remains `3.21.1`). This post-release\nsnapshot adds Chinese-aware title matching and balanced CJK wrapper handling,\ndeclares the Markdown parser dependency used by surface-form checks, makes a\nmissing PyYAML dependency fail visibly, and carries the September\nharness-retirement audit plus documentation corrections. The v3.21.1\nresearch-workflow, inquiry-ledger, promotion-bakeoff, review-criteria, and\nconsent boundaries remain unchanged.\nNested upstream `.github/` workflows and root `agents/` mirrors are preserved\nfor traceability and self-tests, but are not repo-level CI or Codex entrypoints;\nClaude/plugin loader files under `.claude/` and `.claude-plugin/` remain\nintentionally excluded.\n\n## Install ARS-Codex Plugin\n\nAdd the GitHub marketplace and install ARS-Codex with Codex CLI:\n\n```bash\ncodex plugin marketplace add Imbad0202/academic-research-skills-codex --ref main\ncodex plugin add ars-codex@ars-codex\n```\n\nTo update a plugin install later:\n\n```bash\ncodex plugin marketplace upgrade ars-codex\ncodex plugin add ars-codex@ars-codex\n```\n\nIn Codex Desktop, you can alternatively add the repository from **Plugins** and\nthen install **ARS-Codex**:\n\n```text\nMarketplace source: https://github.com/Imbad0202/academic-research-skills-codex.git\nBranch/ref: main\nPlugin: ars-codex\n```\n\nThe plugin root is `plugins/ars-codex/`. Its `skills/` directory contains a\nmaterialized copy of `academic-research-suite`, not a symlink. This keeps\nCodex Desktop installs portable on Windows, where plugin caches may materialize\nsymlinks as plain text files and skip bundled skill registration.\n\nOpen a new Codex conversation after installation, then invoke\n`$academic-research-suite` or describe an academic research task that matches\nthe bundled workflow.\n\n## Direct Skill Install Or Update\n\nAs an alternative to the plugin, install the skill directly from this repo\npath. Use `--method git` so public and\ncredentialed GitHub access both work consistently:\n\n```bash\npython3 \"$HOME/.codex/skills/.system/skill-installer/scripts/install-skill-from-github.py\" \\\n  --repo Imbad0202/academic-research-skills-codex \\\n  --ref main \\\n  --path skills/academic-research-suite \\\n  --method git\n```\n\nOn macOS and many Linux systems, Python 3 is exposed as `python3` rather than\n`python`. If your system only has a `python` command and it is Python 3, use\n`python` in the commands instead.\n\nTo update an existing install:\n\n```bash\nrm -rf \"$HOME/.codex/skills/academic-research-suite\"\npython3 \"$HOME/.codex/skills/.system/skill-installer/scripts/install-skill-from-github.py\" \\\n  --repo Imbad0202/academic-research-skills-codex \\\n  --ref main \\\n  --path skills/academic-research-suite \\\n  --method git\n```\n\nOpen a new Codex conversation after installation. Existing Codex sessions may\nkeep their old skill cache; you do not need to close unrelated Claude or Codex\nsessions.\n\nVerify with `/skills`: you should see one ARS-Codex entry,\n`academic-research-suite` or `ARS-Codex`. You should **not** see separate `academic-paper`,\n`academic-pipeline`, `deep-research`, or `academic-paper-reviewer` skills from\nthis package. If you do, reinstall with the update command above and open a new\nCodex conversation.\n\n## Codex Docs\n\n- [Codex setup](skills/academic-research-suite/ars/docs/SETUP.md) covers\n  installation, `ars-*` aliases, optional tools, Material Passport adapters,\n  and unsupported Claude plugin features.\n- [Codex architecture](skills/academic-research-suite/ars/docs/ARCHITECTURE.md)\n  explains the logical ARS pipeline with the Codex runtime overlay.\n- [Optional full-runtime adapter](CODEX_FULL_RUNTIME_ADAPTER.md) documents the\n  disabled-by-default planner, Codex agent-team templates, and hook pack.\n\n## Usage\n\nInvoke the suite explicitly with `$academic-research-suite` (singular), then\ndescribe the research task and provide any source files, notes, draft text,\nreviewer comments, or output constraints.\n\n```text\nUse $academic-research-suite to help me plan a systematic literature review on\nAI adoption in higher education quality assurance.\n```\n\nThe Codex adapter routes the request to one of five ARS workflows:\n\n| Workflow | Use when you need | Example prompt |\n|---|---|---|\n| `deep-research` | Research question refinement, literature review, systematic review, meta-analysis, fact-checking | `Use $academic-research-suite to build a systematic review protocol for AI in higher education QA.` |\n| `academic-paper` | Paper outline, drafting, abstract, revision, citation formatting, AI disclosure | `Use $academic-research-suite to turn these notes into an IMRaD paper outline and drafting plan.` |\n| `academic-paper-reviewer` | Manuscript review, simulated peer review, editorial decision, re-review | `Use $academic-research-suite to review this manuscript and produce a journal-style decision letter.` |\n| `academic-pipeline` | End-to-end research-to-paper workflow with integrity gates, review, revision, and final checks | `Use $academic-research-suite to run an end-to-end research-to-paper pipeline from topic to revised manuscript.` |\n| `experiment-agent` | Code experiment planning, human study protocol, statistical interpretation, reproducibility validation | `Use $academic-research-suite to plan a code experiment and define reproducibility checks.` |\n\n### Claude-Style Aliases\n\nClaude Code v3.7 installs `/ars-*` slash commands. Codex does not have the same\nplugin command registry, so this package emulates the command intent inside the\nsingle `$academic-research-suite` skill. Use either form:\n\n```text\nUse $academic-research-suite: ars-plan my paper on AI governance in universities.\n```\n\nor, when your Codex client passes slash-prefixed text through as a normal user\nmessage:\n\n```text\n/ars-plan my paper on AI governance in universities.\n```\n\nIf slash input is intercepted by the client, use the plain alias form:\n\n```text\nars-plan my paper on AI governance in universities.\n```\n\n| Claude command | Codex alias | Routed workflow |\n|---|---|---|\n| `/ars-plan` | `ars-plan` | `academic-paper` `plan` mode |\n| `/ars-outline` | `ars-outline` | `academic-paper` `outline-only` mode |\n| `/ars-abstract` | `ars-abstract` | `academic-paper` `abstract-only` mode |\n| `/ars-lit-review` | `ars-lit-review` | `academic-paper` `lit-review` mode |\n| `/ars-citation-check` | `ars-citation-check` | `academic-paper` `citation-check` mode |\n| `/ars-disclosure` | `ars-disclosure` | `academic-paper` `disclosure` mode |\n| `/ars-format-convert` | `ars-format-convert` | `academic-paper` `format-convert` mode |\n| `/ars-revision-coach` | `ars-revision-coach` | `academic-paper` `revision-coach` mode |\n| `/ars-revision` | `ars-revision` | `academic-paper` `revision` mode |\n| `/ars-reviewer` | `ars-reviewer` | `academic-paper-reviewer` full mode |\n| `/ars-mark-read` | `ars-mark-read` | Human-read signal for citation keys in the active Material Passport |\n| `/ars-unmark-read` | `ars-unmark-read` | Rescind a prior human-read signal |\n| `/ars-cache-invalidate` | `ars-cache-invalidate` | Invalidate cached verification entries for one citation key |\n| `/ars-full` | `ars-full` | `academic-pipeline` full workflow |\n\n### Working Pattern\n\nFor best results, start with the workflow goal and the current state of your\nmaterials:\n\n```text\nUse $academic-research-suite.\n\nGoal: write a journal article.\nCurrent materials: I have a literature matrix and rough findings, but no outline.\nOutput needed now: paper architecture and missing-evidence checklist.\nConstraints: English, APA 7, higher education policy audience.\n```\n\nIf you only have a paper topic or broad research direction and do not yet have a\nclear research question, the Codex router should start with ARS Socratic\nscoping:\n\n```text\nUse $academic-research-suite.\n\nI want to write a paper on AI adoption in higher education quality assurance.\nI do not yet have a clear research question.\nPlease use SCR / Socratic dialogue to help me narrow the question first; do not write an outline yet.\n```\n\nExpected route: `deep-research` `socratic` mode first. ARS should ask narrowing\nquestions and should not produce an outline or draft until the research question\nhas converged.\n\nFor review tasks, provide the manuscript or a path to the manuscript, plus the\nreview mode you want:\n\n```text\nUse $academic-research-suite to review this paper.\nMode: full review.\nFocus: methodology, contribution, citation integrity, and likely desk-reject risks.\nOutput: reviewer reports plus editorial decision letter.\n```\n\nFor staged pipelines, ask for a checkpoint instead of asking Codex to run the\nentire process silently:\n\n```text\nUse $academic-research-suite to start an academic-pipeline run.\nBegin with Stage 0 intake and stop after producing the pipeline dashboard.\n```\n\n### Smoke Tests\n\nIn a new Codex conversation:\n\n```text\n/skills\n```\n\nExpected: one ARS entry only.\n\nThen test Socratic routing:\n\n```text\nUse $academic-research-suite.\nI want to write a paper on AI adoption in higher education quality assurance.\nI do not yet have a clear research question.\n```\n\nExpected: route to `deep-research` `socratic` mode and ask narrowing questions.\n\nCLI smoke test:\n\n```bash\ncodex exec --ephemeral --sandbox read-only \\\n  -C /path/to/academic-research-skills-codex \\\n  'Use $academic-research-suite. Router smoke test only. User request to classify: I want to write a paper on AI adoption in higher education quality assurance, but I do not yet have a clear research question. According to the academic-research-suite router, classify the workflow and mode.'\n```\n\nMaintainer quality gates:\n\n```bash\npython3 skills/academic-research-suite/codex/scripts/ars_codex_quality_gates.py all --json\n```\n\nExpected: every reported gate has `\"ok\": true`.\n\n### Non-Blocking Codex Warnings\n\nThese Codex messages do not mean ARS failed to install:\n\n- `[features].codex_hooks is deprecated` — update your Codex config when\n  convenient; ARS-Codex does not require hooks for normal use.\n- `hooks need review before they can run` — review those hooks separately if\n  you use them. ARS-Codex treats vendored Claude hooks as traceability metadata\n  and does not require them.\n\n### Codex Adapter Behavior\n\nARS was originally written for Claude Code. In this Codex package:\n\n- The vendored `agents/*.md` files are used as role and phase prompts.\n- The Codex-only `codex/` directory contains an optional full-runtime adapter\n  profile. It is disabled by default and does not change normal inline routing.\n- The vendored `commands/ars-*.md` files are prompt recipes only. Codex does not\n  register them as slash commands.\n- The vendored `hooks/hooks.json` file is preserved for upstream traceability\n  only. Codex does not install Claude Code hooks from this package.\n- Codex does not automatically spawn background agents unless you explicitly ask\n  for delegated or parallel agent work.\n- Web/source verification uses Codex browsing and must cite sources when current\n  or external facts matter.\n- Cross-model verification is disabled by default. When explicitly requested in\n  this Codex package, follow the vendored provider setup in\n  `ars/shared/cross_model_verification.md`, identify the provider/model/content\n  class first, and obtain explicit user consent before any external upload.\n  External reviewers are called through configured provider APIs, not simulated\n  through the active Codex model.\n- `ARS_MODEL_TIERING` is unset by default. The Codex adapter preserves the\n  upstream judgment/execution classification but applies `economy` or\n  `quality-boost` only when the runtime supports an explicit per-dispatch model\n  override; otherwise it reports a no-op and keeps the active model.\n- Protected top-level agent `tools:` allowlists remain least-privilege role\n  boundaries. A dispatched checkpoint owner does not gain Bash or network\n  transport; the dispatching Codex context owns any explicitly consented\n  cross-model call.\n- A `[CROSS-MODEL-HANDOFF v1]` block is a transport request, not a deliverable.\n  The dispatcher validates it, sends only its payload, applies the mechanical\n  result routing, and returns judgment work to the original owner.\n- In reviewer `full` mode, an explicitly configured and consented cross-model\n  run swaps the existing Reviewer 2 seat; it never adds a sixth reviewer.\n  Re-review applies the separate Priority-1 judge pass and records provenance.\n  Single-family execution and provider fallback are disclosed.\n- `ARS_CACHE_STALE_ADVISORY_DAYS` controls the advisory-only cache-age threshold,\n  while `ARS_CACHE_REVALIDATE=1` opts into live bibliographic re-validation.\n  These settings apply when the programmatic citation gate is run; stale rows\n  alone never fail an integrity gate.\n- Locally read PDFs run the structural `pdf_read_preflight.py` by default before\n  page anchors are trusted. `FAIL` and `UNAVAILABLE` remain distinct, and a\n  missing parser or sidecar is never treated as `PASS`. The v3.20 sandboxed\n  content classifier is an explicit opt-in advisory and cannot override this\n  structural preflight.\n- `ars-mark-read` requires a user-declared `read_scope` for every new mark.\n  Explicit unknown and legacy scope-less records remain `coverage_unknown`;\n  partial coverage stays visible and Codex never infers full-text reading.\n- Revision rounds use the v3.20 evidence-bound, non-ranking roadmap and explicit\n  author-adjudication contract. They also preserve the v3.19-introduced\n  claim-strength ladder and deterministic numeric, citation, marker, and\n  protected-term conservation checks as advisory-first guards.\n- The upstream v3.18 SessionStart update checker is vendored but not installed\n  or executed as a Codex hook. Plugin users update with\n  `codex plugin marketplace upgrade ars-codex` followed by\n  `codex plugin add ars-codex@ars-codex`; direct skill installs still update by\n  reinstalling or pulling this repository.\n- Upstream references to a \"fresh Claude Code session\" mean a new Codex\n  conversation in this package; Material Passport reset semantics still apply.\n- If a citation, source, statistic, or journal policy cannot be verified, Codex\n  should mark it as unverified rather than invent support.\n\n### ARS post-v3.21.1 Main Parity\n\nThis package aims for the same user-facing workflow content as upstream ARS\npost-`v3.21.1` `main` at\n`94436237913091d4739870159d241660527e8338` where Codex has an equivalent\nconcept. Upstream suite metadata remains `3.21.1`.\n\nBibliographic network behavior is intentionally explicit at the Codex adapter\nboundary:\n\n| Research path | Default Codex behavior | Dedicated API/client trigger |\n|---|---|---|\n| Ordinary topic or candidate discovery | Codex browsing and authoritative web sources | The four Python resolver clients are not launched |\n| Prompt-level ingest, deduplication, or source verification | Codex browsing or official metadata pages | “Automatic” lookup wording in vendored prompts does not launch a Python client |\n| Script-backed citation-existence gate | Not implied by `ars-full`; Stage 2.5/4.5 still run as integrity checkpoints through the default Codex route | Explicit programmatic-verification request; then Crossref/OpenAlex/Semantic Scholar run for non-manual references and arXiv only when `arxiv_id` exists, subject to cache behavior |\n| Claim-standing discovery | Advisory offer after an eligible Stage 2.5/4.5 Claim Registry row | Separate user request plus affirmative plan-bound consent; uses v3.21 discovery adapters, not the single-reference resolver clients |\n| Contamination backfill or migration | Never automatic | Explicit migration CLI only |\n\n| Upstream ARS feature | Codex package behavior |\n|---|---|\n| One installable plugin | Native Codex plugin `ars-codex`, bundling the single `academic-research-suite` skill |\n| `/ars-*` slash commands | Emulated as `ars-*` aliases through the skill router; not native slash commands |\n| Four upstream skills auto-discovered from `skills/` symlinks | Single Codex router skill selects the workflow and reads the vendored workflow `WORKFLOW.md` files |\n| Plugin-shipped agents | Agent files are role/phase prompts; Codex runs them inline unless the user explicitly asks for delegated subagents |\n| Optional Codex full-runtime profile | Planner, agent-team templates, and hook pack live under `skills/academic-research-suite/codex/`; disabled by default |\n| Heavy commands (`ars-full`, `ars-reviewer`, `ars-revision-coach`) omit `model:`; light modes retain `model: sonnet` | Heavy commands inherit the current Codex session model; light-mode `sonnet` remains upstream Claude metadata and does not override the session model |\n| `ARS_MODEL_TIERING=economy\\|quality-boost` | Classification is preserved; routing remains advisory unless Codex exposes per-dispatch model selection |\n| Protected agent `tools:` allowlists | Preserved as least-privilege role boundaries; dispatched owners do not receive Bash/network transport |\n| Canonical cross-model handoff envelope | Dispatcher validates the envelope, transports only the payload after consent, and follows the closed result-routing contract |\n| Contained Codex citation transport | Opt-in, consent-gated transport is limited to narrow citation-integrity checks; it is inactive unless explicitly configured and requested |\n| Evidence-bound review and revision | Durable evidence rows, confirmed review criteria, non-ranking roadmaps, author adjudication, and revision-evidence bundles are preserved |\n| Research-workflow profiles | Default-off deterministic selection with a visible field-general fallback; no research family is inferred from manuscript content, and corrections stale rather than rewrite prior artifacts |\n| Inquiry branch ledger | `ARS_INQUIRY_LEDGER=1` enables the local opt-in alpha; author events, bounded summaries, path/lock/recovery safeguards, and stale causes are preserved without granting network authority |\n| Sealed model-promotion bakeoffs | Commitment/reveal contracts and hermetic tests are vendored; the history-dependent tree verifier remains upstream-only because the re-rooted subtree has no complete upstream Git history |\n| Socratic research-question authorship | Non-convergence never triggers system-authored candidate questions; explicit user request is required to leave non-generation mode |\n| Categorical reviewer judgement and panel provenance | Live packages remain `NOT_CALIBRATED`; no numeric score, weight, aggregate, ranking, or binary independence claim is fabricated |\n| Review criteria and human-subjects authority | Venue/criteria and ethics/data-protection authority require explicit user confirmation; Codex does not infer or simulate approval |\n| Optional PDF content classifier | The sandboxed classifier is an opt-in advisory dependency and cannot override structural PDF preflight results |\n| Cross-model Reviewer 2 and re-review judge tracks | Available only with explicit provider configuration and content consent; the fixed seat, Judge Record, single-family disclosure, and fallback disclosure are preserved |\n| Source-backed review-criteria proving set | Exact-profile author confirmation, source receipts, and three-consumer digest binding are preserved; one proving profile is not venue or discipline coverage |\n| Cache staleness advisory and live re-validation | Local cache remains the default; stale rows are advisory-only and `ARS_CACHE_REVALIDATE=1` opts into live bibliographic checks |\n| Data-flow and capability transparency | The v3.21 network map, control-availability matrix, stage-capability matrix, risk register, and governance statement are vendored without promoting evidence labels into effectiveness or certification claims |\n| Claim-standing pipeline wiring | Eligibility only offers the advisory view; query-plan binding, explicit consent, freshness validation, and transmission accounting remain mandatory before any external call |\n| Risk-stratified claim, scope, and novelty checks | Vendored workflow prompts and schemas preserve high-impact-first sampling plus advisory-only scope and search-bounded novelty rows |\n| Local-PDF read-integrity preflight | The structural pypdf preflight and sidecar contract remain the default; parser unavailability or repair warnings stay explicit `UNAVAILABLE` advisories, while the v3.20 classifier above remains opt-in only |\n| Human-read scope attestation | Every new mark requires user-owned `read_scope`; legacy missing scope remains unknown and partial coverage stays distinguishable from full coverage |\n| Claim coverage and bounded evaluation substrates | Exact registered-claim coverage, drift dispositions, claim-standing stance tools, and blind ideation assignment preserve provenance and unmeasured boundaries without certifying semantic completeness or correctness |\n| Revision claim-drift guards | The v3.20 non-ranking roadmap and author-adjudication contract complement the claim-strength ladder, revision-evidence bundle, deterministic token-conservation checker, and held-out measurement set |\n| Executable panel/degradation/pipeline-boundary checks | Vendored with their hermetic tests and exposed by the optional full-runtime manifest |\n| SessionStart and SubagentStop hooks, including the update reminder | Vendored for traceability only; Codex does not install or execute Claude hooks |\n| Plugin marketplace update | Refresh with `codex plugin marketplace upgrade ars-codex`, then re-add `ars-codex@ars-codex`; direct skill installs still reinstall or pull |\n| Claude Code Agent Team | Not automatic; Codex subagents require an explicit user request for delegation or parallel agents |\n| Cross-model provider dispatch from upstream docs | Disabled by default; available only with explicit provider configuration and explicit user consent |\n\n### Optional External Cross-Model Reviewer API\n\nFor reviewer calibration or cross-model devil's advocate checks, configure one\nof the provider tuples documented in\n`ars/shared/cross_model_verification.md`, then ask for cross-model verification\nexplicitly in the prompt. For example:\n\n```bash\nexport OPENAI_API_KEY=\"<your-openai-api-key>\"\nexport ARS_CROSS_MODEL=\"gpt-5.5\"\n```\n\nWithout both a configured provider and explicit user consent for the content\nclass being sent, ARS-Codex falls back to single-runtime review and reports that\ncross-model verification was unavailable.\n\n## Support And Sponsorship\n\nIf ARS-Codex helps your research workflow, you can support maintenance through\n[Buy Me a Coffee](https://buymeacoffee.com/crucify020v).\n\n## Security\n\nDo not open public issues for vulnerabilities. Follow\n[`SECURITY.md`](SECURITY.md) for private reporting, and see the\n[release readiness and security report](security_best_practices_report.md) for\nthe latest local validation summary.\n\n### File Layout For Advanced Use\n\nThe entry point is:\n\n```text\nskills/academic-research-suite/SKILL.md\n```\n\nWorkflow content is under:\n\n```text\nskills/academic-research-suite/ars/<workflow>/\n```\n\nShared schemas, compliance rules, and cross-workflow contracts are under:\n\n```text\nskills/academic-research-suite/ars/shared/\n```\n\nWhen debugging or updating the package, preserve these paths. Many ARS workflow\nfiles cross-reference `shared/`, `scripts/`, `examples/`, and other workflow\ndirectories.\n\n## Update Policy\n\nUpdates sync selected upstream ARS content into `skills/academic-research-suite/ars/`.\nDo not mirror the Claude Code repo blindly; exclude Claude/plugin loader files\nsuch as `.claude/`, `.claude-plugin/`, source `.gitignore`, and symlink-only\nalias directories that are not needed in Codex. Nested upstream `.github/`\nworkflows may be retained as inactive traceability and self-test fixtures.\n\n### Inactive Upstream Scripts\n\nSome upstream maintenance scripts are vendored but intentionally inactive in\nthis Codex package because they require non-vendored Claude Code inputs such as\n`.claude/CLAUDE.md`. See `inactive_upstream_scripts` in\n`skills/academic-research-suite/manifest.json` before wiring any upstream script\ninto Codex CI.\n\n## Contributors And Acknowledgements\n\n**Cheng-I Wu** - Maintainer of the ARS suite and this Codex sibling\ndistribution.\n\n**Codex** - Assisted with the Codex adapter packaging, router-policy hardening,\ntest fixes, and release-readiness review under maintainer direction.\n\n**[vinschger](https://github.com/vinschger)** - Reported beginner installation\nfriction around `python` vs `python3`, which led to clearer setup instructions\nfor macOS and other environments.\n\n**[Joker2377](https://github.com/Joker2377)** - Helped answer community\ninstallation questions and clarify beginner setup steps in issue discussions.\n\nVendored upstream ARS contributors are acknowledged in\n[`skills/academic-research-suite/ars/README.md`](skills/academic-research-suite/ars/README.md#contributors).\n",
  "bytes": 27124,
  "sha": "61dd1fccda46406085812b84bab20ce55bc09c4b83cee6f15d8bee6a49ee97e6",
  "repo_slug": "imbad0202/academic-research-skills-codex",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_imbad0202_academic_research_skills_codex_36c19951/readme"
}