Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Orch Prd

ASecurity

Use when the user wants to turn a product or feature idea into a PRD for an orch-managed project (step 1 of orch-plan). Interviews for the missing facts, then writes docs/prd/NNN-<slug>.md with numbered, testable requirements that orch-arch and orch-spec can trace. Never writes code, specs or tasks.json.

2 stars
0 votes
0 copies
0 views
Added 9/23/2026
ai-agentsgogitsecurityperformance

Works with

cli

Security Analysis

A100/100

Scanned 9/23/2026

Install to Claude Code

$npx -y skills add hectorcanaimero/orch --skill orch-prd --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Orch Prd?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Orch Prd
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/hectorcanaimero-orch-prd/badge)](https://www.skillsdirectory.com/skills/hectorcanaimero-orch-prd)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: orch-prd
description: Use when the user wants to turn a product or feature idea into a PRD for an orch-managed project (step 1 of orch-plan). Interviews for the missing facts, then writes docs/prd/NNN-<slug>.md with numbered, testable requirements that orch-arch and orch-spec can trace. Never writes code, specs or tasks.json.
---

# orch-prd — idea → product requirements

First stage of the orch pipeline: **PRD → ARCH → SPEC → TASKS**. The output is
a document a human signs off on and the next stage (`orch-arch`) builds from.
Nothing here is dispatched to an agent directly, so optimise for a reader who
has to say "yes, that is what we are building" — not for an implementer.

## Before writing

1. Confirm you are in an orch project: `tasks.json` or `.orchestrator/` at the
   root. If neither exists, stop and suggest `orch init` (or
   `orch init --template <name>`); the later stages need its layout.
2. Read what already exists so the PRD does not contradict it: `docs/prd/`,
   `docs/arch/`, `specs/`, the README, and `orch explain` for where the
   project stands.
3. Ask — in **one** batch, not one question per message — for whatever the
   idea leaves open and the PRD cannot honestly guess:
   - who the users are and what they do today instead;
   - the one outcome that makes this worth building;
   - hard constraints (deadline, budget, stack already chosen, compliance);
   - what is explicitly **out** of scope.

   If the user says "just decide", decide, and record each decision under
   *Assumptions* so it is visible and reversible.

## Where it goes

`docs/prd/NNN-<slug>.md` — `NNN` is the next free three-digit number in that
directory (`001` for the first), `<slug>` is short kebab-case.

**Never under `specs/`.** `orch atomize` walks the spec root and warns about
any file whose frontmatter is not `type: spec`; a PRD there is noise in every
atomize run.

## Frontmatter

```yaml
---
type: prd
project_id: <id>  # the first line of `orch explain` reads "Project <id>"
version: 0.1
generated_by: orch-prd
generated_at: <YYYY-MM-DD>
title: <Feature name>
---
```

`type: prd` is one of the values atomize recognises (`prd`, `arch`, `spec`,
`poc`, `task-batch`); any other value is reported as invalid.

## Sections

```markdown
# <Feature name>

## Problem
Two or three paragraphs: who hurts, how, and what it costs them today.

## Users
One bullet per user type, with the job they are trying to get done.

## Goals
- G1 — <measurable outcome>

## Non-goals
- <thing a reasonable reader might expect, and why it is out>

## Functional requirements
- **FR-1** — <the system does X when Y>. *Acceptance:* <observable check>.

## Non-functional requirements
- **NFR-1** — <performance / security / accessibility / cost bound>.

## Assumptions
- <decision taken without the user, so it can be challenged>

## Open questions
- <what blocks a confident architecture, with who can answer it>

## Milestones
- M1 — <user-visible slice>: FR-1, FR-2
```

## Rules

- **Every requirement is numbered and testable.** `FR-n` / `NFR-n` ids are
  what `orch-arch` and `orch-spec` cite; renumbering one later breaks that
  trace, so append new ones instead.
- **Acceptance is observable.** "Works well" is not a criterion; "a signed-out
  user who opens /orders is redirected to /login" is.
- **No implementation.** No frameworks, tables or endpoints unless the user
  imposed them as a constraint — then they go under *Non-functional
  requirements* as a constraint, not as a design.
- **Milestones are user-visible slices**, each naming the requirements it
  delivers. Each one becomes a phase in `orch-spec` (`F1`, `F2`, …), and a
  phase is what orch shows as a milestone — the client portal's and the
  dashboard's progress bars — so a client should recognise every one.
- **Open questions are allowed; silent guesses are not.** A PRD with two
  honest open questions beats one that hides them.

## Hand-off

Show the user the path and a five-line summary (goals, FR count, milestones,
open questions). Stop there. The next step is `orch-arch`, and only once the
user accepts the scope — or `orch-plan` resumes it for them.

Attribution

hectorcanaimerohectorcanaimero
View sourceMore from hectorcanaimero →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

693621 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →