prompt-writing SKILL.md skillA

prompt-writing from mangowhoiscloud/geode is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add mangowhoiscloud/geode/prompt-writing@git:20260920.1df5edb

Pins git:20260920.1df5edb in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
mangowhoiscloud/geode
Path
.agents/skills/prompt-writing/SKILL.md
Versions recorded
4
Latest version
git:20260920.1df5edb, committed Sep 20, 2026
First version recorded
Aug 25, 2026
Size
4,493 bytes
Repository stars
15
License
Apache-2.0

GEODE prompt-writing standards. Use when creating or editing runtime prompts, system prompts, tool/developer instructions, model-facing identity text, prompt templates, or scaffold rules that affect model behavior; especially when avoiding identity assertions such as "You are ...".

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Prompt Writing

Use this skill before editing model-facing prompt text in GEODE.

## Core Standard

Write prompts as metadata, behavior contracts, and task constraints. Avoid direct
identity assertions such as:

- `You are GEODE`
- `You are a helpful assistant`
- `Act as ...`
- `You are currently in ...`

Prefer declarative clauses:

- `Agent: GEODE.`
- `Runtime: self-hosting autonomous execution harness.`
- `Mode: task execution.`
- `Voice: direct, concise, operator-facing.`
- `Scope: the user's authorized task.`
- `Completion: report verified results and unresolved limits.`

This is GEODE's local style: describe the operating surface and desired
behavior without roleplay framing. It is not a vendor prohibition on role
prompts. Claude's [official guidance](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices)
uses clear instructions and descriptive XML boundaries, and also permits role
framing. Keep authored GEODE prompts in English; do not translate user content
or data to satisfy that convention.

## Workflow

1. Identify whether the text is model-facing.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add mangowhoiscloud/geode/prompt-writing@sha256:3ac8354e540675d8

The line at the top of this page pins the label git:20260920.1df5edb, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_ai42pesytp2d64uw.svg)](https://markdownregistry.com/a/art_ai42pesytp2d64uw)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 4 versions of this file, between Aug 25, 2026 and Sep 20, 2026, about one every 9 days.

versioncommittedcommitsizeaudit
git:20260920.1df5edb latest2026-09-20 1df5edb 4,493 BA view · diff
git:20260920.d0cb4f12026-09-20 d0cb4f1 2,635 BA view · diff
git:20260914.b6e3b892026-09-14 b6e3b89 2,589 BA view · diff
git:20260825.9516eab2026-08-25 9516eab 2,235 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (4493 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

GitHub

mangowhoiscloud/geode · 15 stars · license Apache-2.0 · pushed 2026-11-02 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_ai42pesytp2d64uw
GET https://markdownregistry.com/api/v1/resolve?ref=mangowhoiscloud/geode/prompt-writing
GET https://markdownregistry.com/api/v1/blob/3ac8354e540675d83d10bc7f57d1daae4efc3d05013318cc789dc726da09f480

Your AI assistant 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.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from mangowhoiscloud/geode

agent-anti-pattern SKILL.md skill
mangowhoiscloud/geode · .agents/skills/agent-anti-pattern/SKILL.md · Audit agent runtimes and their tests for lifecycle inference, capability overload, context pollution, correlated…
git:20260819.789a6b6 · audit A · 15 stars
agent-ops-debugging SKILL.md skill
mangowhoiscloud/geode · .agents/skills/agent-ops-debugging/SKILL.md · Diagnose agent operational failures through default-policy checks, multi-gap analysis, explicit service ownership…
git:20260906.dc9349e · audit A · 15 stars
agent-world-benchmark SKILL.md skill
mangowhoiscloud/geode · .agents/skills/agent-world-benchmark/SKILL.md · Align, run, audit, and publish GEODE benchmark comparisons against Agent-World arXiv 2604.18292v1 across MCP-Mark, BFCL…
git:20260825.9516eab · audit A · 15 stars
anti-deception-checklist SKILL.md skill
mangowhoiscloud/geode · .agents/skills/anti-deception-checklist/SKILL.md · Verification checklist to prevent fake success. Detects test deletion/disabling, coverage regression, lint bypass…
git:20260906.dc9349e · audit B · 15 stars
architecture-patterns SKILL.md skill
mangowhoiscloud/geode · .agents/skills/architecture-patterns/SKILL.md · Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven…
git:20260907.97cb78b · audit A · 15 stars
baseline-epoch-partition SKILL.md skill
mangowhoiscloud/geode · .agents/skills/baseline-epoch-partition/SKILL.md · Content-addressed baseline-archive epoch partitioning. Partition baseline_archive.jsonl + the self-improving hub's…
git:20261101.afbafb2 · audit A · 15 stars
code-review-quality SKILL.md skill
mangowhoiscloud/geode · .agents/skills/code-review-quality/SKILL.md · Python code quality review lens. SOLID principles, dead code, exception handling, resource leaks, thread safety…
git:20260925.c2ce538 · audit A · 15 stars
codebase-audit SKILL.md skill
mangowhoiscloud/geode · .agents/skills/codebase-audit/SKILL.md · Codebase audit + refactoring workflow. Dead code detection, God Object splitting, duplicate function removal, design…
git:20260925.c2ce538 · audit A · 15 stars
codex-mcp-verify SKILL.md skill
mangowhoiscloud/geode · .agents/skills/codex-mcp-verify/SKILL.md · Use an available Codex MCP integration as a read-only second-opinion verifier for GEODE code reviews, GAP audits, and…
git:20260825.9516eab · audit A · 15 stars
dependency-review SKILL.md skill
mangowhoiscloud/geode · .agents/skills/dependency-review/SKILL.md · Review GEODE dependency direction, composition boundaries, circular imports, and optional-dependency loading when…
git:20260906.dc9349e · audit A · 15 stars
explore-reason-act SKILL.md skill
mangowhoiscloud/geode · .agents/skills/explore-reason-act/SKILL.md · Mandatory 3-phase explore-reason-act workflow before code modification. Enforces read-before-write, root cause…
git:20260906.dc9349e · audit A · 15 stars
frontier-harness-research SKILL.md skill
mangowhoiscloud/geode · .agents/skills/frontier-harness-research/SKILL.md · Research process for comparing frontier harnesses (Claude Code, Codex, OpenClaw, autoresearch, Prime Agent) with the…
git:20261101.afbafb2 · audit A · 15 stars

Every file in mangowhoiscloud/geode

Browse by kind, by grade A, or by owner.