prompt

90 / 100 · Low risk

Fine to try.

By ryanthedev/oberskills (GitHub) · design · 64 stars · scanned at commit 746b3c0

Design and review prompts Claude-first — system prompts, reusable agent definitions, long or novel dispatch briefs, pipeline stages, and prompts users will run elsewhere. Two modes: design and adversarial review. Covers Claude model behavior and migration, de-prompting, few-shot design, output-schema ordering, context engineering, prompt security, optimization, non-Claude/Codex target review via references/porting.md in this skill directory, and a verbatim behavior-snippet library. Use when wri…

Works with any agent that supports SKILL.md skills.

What this skill can do

Connects to the internet

The author does not mention this.

Found in 4 places.

mcp/src/lib/agent.ts, line 210...(args.allowNetwork ? {} : { disallowedTools: ["WebSearch", "WebFetch"] }),
mcp/src/tools/run-eval.ts, line 49"WebSearch/WebFetch disallowed unless allow_network, " +
mcp/src/tools/run-eval.ts, line 81allow_network: z.boolean().default(false).describe("When false, WebSearch/WebFetch are disallowed in subject sessions."),

Show 1 more

mcp/src/tools/run-eval.ts, line 87"WebSearch/WebFetch stay disallowed unless allow_network.",

Runs commands on your computer

The author does not mention this. It runs automatically when the plugin is installed.

Found in 6 places.

.claude-plugin/plugin.json, line 12"command": "bun",
.claude-plugin/plugin.json, line 16"command": "bun",
.claude-plugin/plugin.json, line 26"command": "command -v bun >/dev/null 2>&1 || exit 0; diff -q \"${CLAUDE_PLUGIN_ROOT}/mcp/package.json\" \"${CLAUDE_PLUGIN_DATA}/package.json\" >/dev/null 2>&1

Show 3 more

.claude-plugin/plugin.json, line 30"command": "command -v bun >/dev/null 2>&1 || exit 0; diff -q \"${CLAUDE_PLUGIN_ROOT}/mcp-browser/package.json\" \"${CLAUDE_PLUGIN_DATA}/mcp-browser/package.jso
.mcp.json, line 4"command": "bash",
.mcp.json, line 8"command": "bash",

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'prompt' ] || [ -L ~/.agents/skills/'prompt' ]; then echo 'A skill folder named prompt already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/ryanthedev/oberskills' "$d" && git -C "$d" checkout --quiet '746b3c0e4e57c5872df67d8b9d865d38cca86e6a' -- 'skills/prompt' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/prompt' ~/.agents/skills/'prompt' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'prompt' ~/.claude/skills/'prompt' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for ryanthedev/oberskills shows the score of its lowest-scoring skill.

SkillTrust badge for ryanthedev/oberskills

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/ryanthedev/oberskills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/ryanthedev/oberskills/)

Score history

Date	Version	Score
2026-11-08	`746b3c0`	90