Skip to content

Humanities Superpowers — English Social Launch Copy

Replace [REPOSITORY_URL] and [PAGES_URL] with the live links before publication.

1. Main announcement — LinkedIn, Facebook, research communities

Introducing Humanities Superpowers

Generative AI can produce prose that looks like scholarship within seconds. But scholarship is not primarily a text-generation problem. It requires researchers to decide what can responsibly be claimed, verify evidence, distinguish interpretation from fact, define conceptual boundaries, and remain accountable for the final argument.

I created Humanities Superpowers in response to that problem.

Humanities Superpowers is not a tool that writes papers for researchers. It is an open-source scholarly judgment scaffold designed to reduce the risks of rushed claims, fabricated citations, conceptual drift, unsupported generalization, and premature declarations of completion when humanities scholars work with AI agents.

The repository includes:

  • 13 core research skills, from formulating a question to pre-submission verification
  • 1 Level 3 router that diagnoses the current state and selects the smallest valid workflow
  • quality gates using PASS, CONDITIONAL PASS, and FAIL
  • concept-lineage mapping, close reading, argument stress testing, citation auditing, terminology control, manuscript review, and peer-review response
  • rollback to an earlier research stage when a downstream failure reveals an upstream problem
  • tested installation routes for Claude Code and OpenAI Codex, plus not-yet-verified Cursor guidance
  • an English white paper, bilingual documentation, worked examples, schemas, and automated validation

The project is guided by one sentence:

Not an AI paper writer. A scholarly judgment scaffold.

Humanities Superpowers was inspired by obra/superpowers, created by Jesse Vincent and Prime Radiant. It adapts the idea of composable agent skills from software development to humanities research, while independently developing workflows for conceptual genealogy, interpretation, argumentation, citation verification, and peer-review revision.

The framework does not guarantee truth, originality, citation accuracy, or publication. Its purpose is narrower and more defensible: to keep uncertainty visible, prevent unverified material from silently becoming accepted evidence, and preserve the researcher's responsibility for scholarly decisions.

Repository: [REPOSITORY_URL]
Documentation: [PAGES_URL]

Researchers, graduate students, digital-humanities practitioners, and AI-agent users are invited to test, critique, and contribute.

HumanitiesSuperpowers #Humanities #GenerativeAI #DigitalHumanities #ResearchMethods #ScholarlyWriting #AIAgents #ClaudeCode #Codex

2. Short announcement — X, Threads, Bluesky

I am releasing Humanities Superpowers: an open-source research workflow for humanities scholars using AI agents.

It does not write papers for you. It connects research questions, conceptual lineage, close reading, argument design, citation audit, peer-review response, and submission checks through explicit quality gates and rollback.

Not an AI paper writer. A scholarly judgment scaffold.

[REPOSITORY_URL]

HumanitiesSuperpowers #DigitalHumanities #AIAgents

3. Very short link post

Humanities Superpowers is an open-source skill framework for humanities researchers using AI agents. It makes questions, evidence, concepts, arguments, citations, and submission claims pass through explicit research quality gates.

[REPOSITORY_URL]

4. Academic mailing list or association post

I would like to share Humanities Superpowers, an open-source framework for structuring the use of AI agents in humanities research.

Rather than automating paper writing, the project supports research judgment across question formulation, scope control, conceptual lineage, literature dialogue, close reading, argument planning, counterargument, citation auditing, terminology consistency, manuscript review, peer-review response, and final submission verification.

Unverified citations and unsupported claims are preserved as failures rather than polished into apparently complete scholarship. When a failure originates upstream, the research orchestrator routes the project back to the earliest relevant stage.

The repository has been tested with Claude Code and OpenAI Codex. It also provides Cursor installation guidance that has not yet been independently verified. The project is released under the MIT License.

Repository: [REPOSITORY_URL]
Documentation: [PAGES_URL]

Critical feedback and discipline-specific contributions are very welcome.

5. Graduate-student-oriented post

AI can generate a polished paragraph long before a research question, source base, or argument is actually ready.

Humanities Superpowers is an open-source framework that helps graduate students and researchers pause at the right moments:

  • Is the research question genuinely arguable?
  • Has the concept's lineage and boundary been checked?
  • Does a source merely exist, or does it support the claim being made?
  • Is a local reading being generalized beyond the evidence?
  • Did the manuscript actually change before the reviewer response says it did?
  • Is the submission package genuinely ready?

The repository includes 13 core research skills, 1 Level 3 router, explicit quality gates, and worked examples.

[REPOSITORY_URL]

6. Developer and AI-agent community post

Coding agents increasingly rely on structured planning, tests, debugging, and verification. Humanities research with AI agents is still often managed through ad hoc prompts.

Humanities Superpowers explores a different approach: composable research skills governed by contracts, epistemic states, quality gates, a research state machine, and rollback.

Included:

  • 13 core research skills at Level 2
  • 1 Level 3 router
  • research-object, gate-report, and session schemas
  • deterministic conformance and integration tests
  • tested Claude Code and OpenAI Codex routes, plus unverified Cursor guidance

[REPOSITORY_URL]

7. Personal first-project announcement

This is my first public GitHub project.

While using generative AI in humanities research and teaching, I became convinced that its main danger is not poor writing. The more serious risk is fluent output that makes unfinished reasoning, unverified sources, and unresolved conceptual decisions appear complete.

I therefore wanted to build something different from a better paper-writing prompt: a workflow that makes AI-assisted research slower at the moments where scholarly judgment matters most.

The result is Humanities Superpowers.

It includes 13 core research skills and 1 Level 3 router covering the path from research-question formulation to citation auditing, manuscript review, peer-review revision, and submission verification. I hope it can grow as a public methodology that researchers use, criticize, and improve together.

[REPOSITORY_URL]

8. First follow-up after launch

Thank you for the early responses to Humanities Superpowers. The project will not grow by adding features for their own sake. Future changes will be driven by documented failures and real research use.

Feedback is especially valuable on:

  • where the workflow was most or least useful
  • unclear skill triggers or handoffs
  • gates that are too strict or too permissive
  • discipline-specific assumptions that should be made explicit
  • installation or loading issues in Claude Code, Codex, or Cursor

Please use Issues or Discussions in the repository:

[REPOSITORY_URL]