Is the Summary Discipline skill free?

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

What is the Summary Discipline SkillProof Score?

8.0/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 6/10, docs & honesty 5/5.

How do I install Summary 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 Summary 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.

Summary Discipline

Faithful summaries: verbatim numbers, preserved hedges, flagged contradictions.

by SkillProof · Skillproofdev/summary-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 · 93e54f2

Our own skill — tested under the same protocol, disclosed. A pre-registered 6-of-11-document subset benchmark shows the skill cut critical omissions from 10 to 2 and flagged all 3 planted contradictions the baseline silently resolved, while both arms tied 0-0 on added claims, distorted numbers, and attribution losses. Caveats: only 6 of 11 planned documents have been run, and skill outputs are ~1.7x longer than baseline, which partly confounds the omission-reduction gain. 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 6/10
Docs & honesty 5/5

What Summary Discipline does

Turns "summarize this" into an auditable process — every sentence traced to a source passage, numbers copied verbatim, hedges and attribution preserved, and contradictions flagged instead of silently resolved. Built after finding that zero of 174 indexed summarizer skills enforce any checkable rule about the summary-source relationship, only structure and brevity.

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

Without the skill

Baseline summary of a memo with 3 planted numeric contradictions states one budget ($2.9M), one cutover date (April 14), and one headcount (8 engineers) as settled fact — silently picking one side of each conflict without ever noting the memo's own exec summary gives different figures ($2.4M / March 3 / 6 engineers).

With Summary Discipline

Skill-loaded summary flags all three contradictions in a visible Flagged block with both conflicting values intact, and separately preserves reversing caveats the baseline dropped while keeping the headline — e.g. a research abstract's "null result should not be interpreted as evidence of safety."

How to install Summary Discipline

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

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

Commands — how to trigger Summary Discipline

/summary-discipline Faithful summaries: verbatim numbers, preserved hedges, flagged contradictions.

It also activates on plain-language prompts like these:

Summarize this 12-page incident postmortem for the exec team.
Give me the tl;dr of this earnings call transcript.
What does this 40-message Slack thread say? I don't have time to read it all.

Frequently asked questions

Is the Summary Discipline skill free?: Yes. The skill itself is free from Skillproofdev/summary-discipline. SkillProof publishes the install command and an independent test verdict at no cost.
Does Summary 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. A pre-registered 6-of-11-document subset benchmark shows the skill cut critical omissions from 10 to 2 and flagged all 3 planted contradictions the baseline silently resolved, while both arms tied 0-0 on added claims, distorted numbers, and attribution losses. Caveats: only 6 of 11 planned documents have been run, and skill outputs are ~1.7x longer than baseline, which partly confounds the omission-reduction gain. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
What is the Summary Discipline SkillProof Score?: 8.0/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 6/10, docs & honesty 5/5.
How do I install Summary 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 Summary 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.