Is the Run History Skill Builder skill free?

Yes. The skill itself is free from dongshuyan/compass-skills. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Run History Skill Builder SkillProof Score?

9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.

How do I install Run History Skill Builder?

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 Run History Skill Builder 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.

Run History Skill Builder

Turns a completed run or repeated workflow into a validated reusable skill package

by dongshuyan · dongshuyan/compass-skills

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Jul 21, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 750630d

Cloned the repo and confirmed all body-referenced files exist (5 references, scripts/validate_skill_package.py, evals/evals.json; spot-checked 3 via raw HTTP, all 200). Ran the bundled validator on the skill's own package: "validation passed", exit 0. For the output test I built two packages for one task (image thumbnail + EXIF-strip skill): a good-faith BASELINE ignoring the body, and one following the body's Design Rules/Validation gates. The validator flagged the baseline with 4 errors (per-skill README, extra frontmatter keys author/version, non-kebab name, name/dir mismatch) plus a machine-path leak warning (/Users/...), while the skill-following package passed clean, exit 0 — the design rules map directly to enforceable checks. Frontmatter is exactly name+description; no curl|sh, base64, or exfiltration smells (the validator itself scans for private paths). Category corrected from "design" to "coding": it is a developer meta-tool that emits skill packages.

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

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 5/5

What Run History Skill Builder does

A meta-skill that converts real run history (a finished task, browser flow, log trace, or repeatedly refined workflow) into a new reusable skill package or a plan-only skill design. It triggers when the user asks to make a new skill from a conversation or logs, extract a reusable workflow, or produce a design plan before writing files; it explicitly defers upgrading an existing skill to a separate upgrader and refuses to execute the underlying business workflow. Ships a working Python validator that enforces frontmatter shape, file references, privacy-path scanning, and eval schema.

How to install Run History Skill Builder

git clone --depth 1 https://github.com/dongshuyan/compass-skills.git /tmp/run-history-skill-builder-src
mkdir -p ~/.claude/skills
cp -R /tmp/run-history-skill-builder-src/skills/run-history-skill-builder ~/.claude/skills/run-history-skill-builder
# No external deps. Bundled validator needs Python 3 (python3/python/py -3):
#   python3 ~/.claude/skills/run-history-skill-builder/scripts/validate_skill_package.py <target-skill-dir>
# Companion skill in same repo: skills/run-history-skill-upgrader (for upgrading existing skills).

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

Commands — how to trigger Run History Skill Builder

/run-history-skill-builder Turns a completed run or repeated workflow into a validated reusable skill package

It also activates on plain-language prompts like these:

Turn this browser flow into a reusable skill
Extract a skill from my recent conversation logs
Build a skill-design plan from this run history

Frequently asked questions

Is the Run History Skill Builder skill free?: Yes. The skill itself is free from dongshuyan/compass-skills. SkillProof publishes the install command and an independent test verdict at no cost.
Does Run History Skill Builder work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Tested · Works. Cloned the repo and confirmed all body-referenced files exist (5 references, scripts/validate_skill_package.py, evals/evals.json; spot-checked 3 via raw HTTP, all 200). Ran the bundled validator on the skill's own package: "validation passed", exit 0. For the output test I built two packages for one task (image thumbnail + EXIF-strip skill): a good-faith BASELINE ignoring the body, and one following the body's Design Rules/Validation gates. The validator flagged the baseline with 4 errors (per-skill README, extra frontmatter keys author/version, non-kebab name, name/dir mismatch) plus a machine-path leak warning (/Users/...), while the skill-following package passed clean, exit 0 — the design rules map directly to enforceable checks. Frontmatter is exactly name+description; no curl|sh, base64, or exfiltration smells (the validator itself scans for private paths). Category corrected from "design" to "coding": it is a developer meta-tool that emits skill packages.
What is the Run History Skill Builder SkillProof Score?: 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Run History Skill Builder?: 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 Run History Skill Builder 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.