Is the Research Discipline skill free?

Yes. The skill itself is free from Skillproofdev/research-discipline. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Research Discipline SkillProof Score?

8.0/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 7/10, docs & honesty 4/5.

How do I install Research Discipline?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Research Discipline with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Research Discipline

Citation and independent-sourcing discipline for any research-shaped Claude answer.

by SkillProof · Skillproofdev/research-discipline

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 8.0/10
Tested: Jul 11, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 0ebe065

Our own skill — tested under the same protocol, disclosed. Six research questions (12 answers total) were judged claim-by-claim against live primary sources: the baseline agent produced 7 of 54 wrong/stale claims (13.0%), the skill-equipped agent produced 3 of 56 (5.4%) — a 59% relative cut in factual error rate for about +10% tokens. The README also discloses two real regressions (an incomplete pricing comparison, a repeated wrong version number), which we weighed in scoring rather than ignoring. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 4/5
Output vs. baseline 7/10
Docs & honesty 4/5

What Research Discipline does

Enforces eight evidence rules on research-shaped answers — live sources over memory, a citation on every claim, two independent sources for load-bearing facts, verbatim numbers, and an adversarial self-check before delivery. Benchmarked at -59% wrong/stale claims versus a bare agent, with the two cases where it still fell short published alongside the wins.

Research Discipline before / after — same task, with and without

Without the skill

Asked for the current Claude model catalog, the baseline agent answered from a cached table without a live check — and got three context-window figures wrong.

With Research Discipline

With the skill loaded, the agent fetched the live docs page, matched all 11 claims against it, and flagged its own answer [single-source] because every figure traced back to one vendor page.

How to install Research Discipline

git clone https://github.com/Skillproofdev/research-discipline ~/.claude/skills/research-discipline

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Research Discipline

/research-discipline Citation and independent-sourcing discipline for any research-shaped Claude answer.

It also activates on plain-language prompts like these:

What's the latest stable version of PostgreSQL and when did it release?
Compare Stripe vs Paddle pricing for a SaaS doing $50k MRR.
Is it true that the EU AI Act bans all facial recognition technology?

Frequently asked questions

Is the Research Discipline skill free?: Yes. The skill itself is free from Skillproofdev/research-discipline. SkillProof publishes the install command and an independent test verdict at no cost.
Does Research Discipline work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Works. Our own skill — tested under the same protocol, disclosed. Six research questions (12 answers total) were judged claim-by-claim against live primary sources: the baseline agent produced 7 of 54 wrong/stale claims (13.0%), the skill-equipped agent produced 3 of 56 (5.4%) — a 59% relative cut in factual error rate for about +10% tokens. The README also discloses two real regressions (an incomplete pricing comparison, a repeated wrong version number), which we weighed in scoring rather than ignoring. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
What is the Research Discipline SkillProof Score?: 8.0/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 7/10, docs & honesty 4/5.
How do I install Research Discipline?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Research Discipline with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.