Bad Research

michael jackson bad

A keyless deep-research agent that runs as a Claude Code skill — a fork-and-enhance of hyperresearch. It searches wide, filters garbage, grounds every claim to a source, and needs zero API keys: the Claude Code host model supplies all inference, exactly like hyperresearch. Optional local CLIs and a [local] neural extra are enhancements, never requirements.

Install

Bad Research is a small CLI that registers itself as a Claude Code skill. No API keys. Requires Python 3.11–3.13.

# Install the CLI (pipx or uv — either works)
pipx install bad-research
uv tool install bad-research

# Register the /bad-research skill into ~/.claude
bad install

# Verify
bad doctor

bad install writes the entry skill to ~/.claude/skills/bad-research/; the per-step skills install lazily on first use. For a project-local install instead of global, run bad install --project inside the project. bad doctor shows what's wired (host model, keyless search/browse, the optional external CLIs it can drive, the [local] neural stack).

Use it in Claude Code

After bad install, open Claude Code in any project and either:

• Invoke it directly — type the slash command with your question:

  /bad-research Is open-source AI more dangerous than closed-source for national security?
  

• Let Claude trigger it — just ask a research-shaped question ("write me a cited report comparing vector databases", "literature review on GLP-1 drugs") and Claude loads the skill automatically.

It scales to the question: a simple lookup gets a fast cited answer in minutes; a broad or contested one runs the full adversarially-reviewed pipeline (~1.5–2.5 h). The final report and every fetched source land in a vault under ./research/ that compounds across sessions.

Pick the depth (it auto-scales, or force it)

By default the skill auto-routes — a simple, bounded question takes the fast route (a quick cited answer, minutes); a broad or contested one takes the full adversarially-reviewed pipeline (~1.5–2.5 h). You can steer it:

• Want a thorough report without the multi-hour wait? The fast route is the sweet spot — its breadth branch fans out K parallel researchers over a wide multi-source browse, then writes a sectioned, fully-cited answer in minutes. Force it with bad route --apply --fast if the auto-router picked full and you want the quicker take. If you're just trying Bad Research out, start here.
• Dial the effort with --effort minimal|low|medium|high to nudge the route and per-step fan-out (minimal/low bias toward fast; medium/high toward full).

On an interactive run the skill announces the chosen route and its rough ETA before it commits to a long job (and for full it shows the editable plan first), so you're never surprised by a 2-hour job you didn't want. The route is decided from the step-1 decomposition and shown by that up-front in-skill route announcement — so you see which route a query takes before any long work starts.

Want the latest unreleased build? Install from source: pipx install git+https://github.com/LeventySeven/badresearch.git

Updating

Already installed? Upgrade the CLI and re-register the skill so both are current:

# From PyPI (pipx or uv — whichever you installed with)
pipx upgrade bad-research      # or: uv tool upgrade bad-research
bad install                    # refresh the /bad-research skill + agents in ~/.claude

# ...or track the latest source
pipx install --force git+https://github.com/LeventySeven/badresearch.git
bad install

bad install is idempotent — re-run it any time after upgrading the CLI to pull the newest entry skill + agents (the per-step skills refresh lazily on the next /bad-research run). Confirm with bad --version.

What it does

A tier-adaptive pipeline turns a question into an audited, fully-cited report, and every fetched source lands in a persistent, searchable vault that compounds across sessions. Keyless by design:

• Search — the host WebSearch tool + DuckDuckGo + 7 scholarly APIs, fused and reranked by the host model.
• Content — a native fetch-and-clean pipeline (readability → markdown → optional LLM clean), SSRF-guarded.
• Browse — an agentic observe → act → extract loop driven by a local, keyless headless browser.
• Retrieve — SQLite FTS5/BM25 by default (no model required), with an optional local neural lane.
• Ground — every factual sentence must carry a source citation, and a deterministic ship-gate blocks any uncited claim. Fabricated quotes are caught for free by a byte-identity check; the harder paraphrase-faithfulness cases are judged by the host model (an optional [local] cross-encoder upgrades this to NLI).

How it works & where the patterns came from

Bad Research takes hyperresearch as its base and enhances each stage with patterns drawn from the best deep-research systems — Perplexity, Gemini, Firecrawl, Stagehand, AgentQL, and others — reimplemented to run keyless on the host model. The full write-up, stage by stage with provenance, is in docs/HOW_IT_WORKS.md.

MIT licensed.