Name: prompt-context-design
Author: DevGuyRash

Imported from DevGuyRash/agent-tooling (plugins/agentic-design-and-evaluation/skills/prompt-context-design/SKILL.md). Install upstream with npx skills add DevGuyRash/agent-tooling --skill prompt-context-design. Copyright stays with the author.

Prompt and Context Design

The authoring guidance applies to the text you deliver. Briefs and context covers briefs for delegated agents, prompts for continuing and unattended work, and what a recipient actually receives.

Whether text helps is a question about behavior. You SHALL NOT claim that instructions work or improve behavior unless you observed it: a scenario that exercises the concern, repeated enough to tell the result from luck, against the same scenario without the text, using the trial runtime. Otherwise you SHALL call them untested.

# after Install: the listing is in your workspace's routing pool - a plan picks it for its slot curl -s -X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'

Install prompt-context-design and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "ocm": "1", "id": "devguyrash-agent-tooling-prompt-context-design", "kind": "skill", "name": "prompt-context-design", "description": "Write or revise instructions another AI will follow, such as prompts, skills, AGENTS.md or CLAUDE.md files, briefs for delegated agents, and prompts for recurring or unattended work.", "publisher": "DevGuyRash", "version": "1.0.0", "capabilities": { "domains": [ "general" ], "tags": [ "skill-md", "github" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Write or revise instructions another AI will follow, such as prompts, skills, AGENTS.md or CLAUDE.md files, briefs for delegated agents, and prompts for recurring or unattended work." ], "primary": false, "metadata": { "source": { "provider": "github", "repository": "https://github.com/DevGuyRash/agent-tooling", "path": "plugins/agentic-design-and-evaluation/skills/prompt-context-design/SKILL.md", "ref": "603d229115a665ddc202ade8b897d6cfc05279a3", "url": "https://github.com/DevGuyRash/agent-tooling/blob/603d229115a665ddc202ade8b897d6cfc05279a3/plugins/agentic-design-and-evaluation/skills/prompt-context-design/SKILL.md", "key": "DevGuyRash/agent-tooling/plugins/agentic-design-and-evaluation/skills/prompt-context-design/SKILL.md" }, "compatibility": "Requires the complete Agentic Design & Evaluation plugin, including its authoring guidance and the Split Testing trial runtime." }, "instructions": "# Prompt and Context Design\n\nThe [authoring guidance](../foundational-knowledge/references/governing-architecture.md) applies to the text you deliver. [Briefs and context](references/briefs-and-context.md) covers briefs for delegated agents, prompts for continuing and unattended work, and what a recipient actually receives.\n\nWhether text helps is a question about behavior. You SHALL NOT claim that instructions work or improve behavior unless you observed it: a scenario that exercises the concern, repeated enough to tell the result from luck, against the same scenario without the text, using th", "cost": { "context_tokens": 174 } }

prompt-context-design

About

Prompt and Context Design

Use it

Install prompt-context-design and it becomes one more candidate the router can pick - when it fits.

Manifest

Reviews