prompt-design skillA
prompt-design is agent-read markdown (skill) from fmind/dot: Design AI application prompts: instructions, context, tools, and output contracts..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Prompt Design Design production prompt stacks with explicit instruction precedence, trusted context, tool contracts, and measurable output behavior. [task-prompts](../task-prompts/SKILL.md) owns task and continuation prompts. ## Workflow 1. **Define the behavior**: user goal, input/output contract, model/version, budget, failure modes, and observed baseline. 1. **Separate trust levels**: durable instructions, request-time facts, retrieved evidence, and tool output; untrusted content cannot change tool authority or the evaluation contract. 1. **Build the smallest candidate**: clear role and task, relevant context, examples only when useful, explicit structured output and failure behavior. 1. **Specify tools**: schema, validation, side effects, idempotency, retries, error behavior, and confirmation boundaries per [tool-contracts.md](references/tool-contracts.md). 1. **Version and evaluate**: prepare [prompt-candidate.md](references/prompt-candidate.md), test on development cases, and use [agent-evaluation](../agent-evaluation/SKILL.md) for the comparison protocol and the project or provider's runner for execution before changing production. ## Gotchas …
Read the whole file at its exact version.
How to install
mdr add fmind/dot/prompt-design@git:20260919.4491844mdr add fmind/dot/prompt-design@sha256:eed627be9a5ad2a3Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_akms5wgiwljwv4ky)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260919.4491844 latest | 2026-09-19 | 4491844 | 3,143 B | A | view · diff |
| git:20260916.50d6810 | 2026-09-16 | 50d6810 | 3,126 B | A | view · diff |
| git:20260910.54ed32c | 2026-09-10 | 54ed32c | 3,154 B | A | view · diff |
| git:20260909.892d859 | 2026-09-09 | 892d859 | 3,194 B | A | view · diff |
| git:20260906.ede7325 | 2026-09-06 | ede7325 | 3,142 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (3143 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
fmind/dot · 9 stars · license MIT · pushed 2026-09-24 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_akms5wgiwljwv4ky GET https://markdownregistry.com/api/v1/resolve?ref=fmind/dot/prompt-design GET https://markdownregistry.com/api/v1/blob/eed627be9a5ad2a34581ed10c7eaaa5ce3562b09334db865ab4ffb5d2e0f2528
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.