Skip to content
Back to skills

Sdd Artifacts

ASecurity

Use when editing tracked SDD artifacts that require naming, EARS, traceability, evidence, or status conventions.

  • 2 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 4, 2026
ai-agentsgit

Security analysis

A100/100

Scanned September 4, 2026

npx -y skills add paulasilvatech/awesome-harness-primitives --skill sdd-artifacts --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Sdd Artifacts?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Sdd Artifacts
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/paulasilvatech-sdd-artifacts/badge)](https://www.skillsdirectory.com/skills/paulasilvatech-sdd-artifacts)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: sdd-artifacts
description: >-
  Use when editing tracked SDD artifacts that require naming, EARS, traceability, evidence, or
  status conventions.
paths:
  - ".specs/**/*.md"
  - ".specs/**/*.yaml"
  - ".specs/**/*.json"
user-invocable: false
---

<!-- Generated from harness/github-copilot/plugins/open-horizons-platform/instructions/sdd-artifacts.instructions.md by harness/claude-code/scripts/convert_from_copilot.py. Edit the source, not this file. -->

# Specification-Driven Development Artifacts

The `sdd-spec-engineer` skill owns lifecycle and generation procedures. These instructions own only the durable shape of files under `.specs/`.

## Conventions

- Name feature directories with a zero-padded numeric prefix and kebab-case slug, matching the `feature_id` in artifact frontmatter.
- Use the established uppercase artifact names such as `SPECIFICATION.md`, `DESIGN.md`, `TASKS.md`, `SOURCE_TRACEABILITY.md`, `VERIFICATION.md`, and `DECISIONS.md`.
- Give every normative requirement a unique stable ID and one observable EARS response using `SHALL`; do not reuse retired IDs for different behavior.
- Give acceptance criteria stable IDs tied to their parent requirement.
- Preserve bidirectional traceability from source and decision through requirement, design, task, test, verification, and evidence.
- Distinguish requested, brownfield, official, repository-constraint, and greenfield claims; greenfield behavior requires an explicit decision.
- Treat live-state claims as dated evidence with source, scope, result, and freshness limits. Plans and expected output are not proof of execution.
- Keep status, implementation state, blockers, and verification results truthful; use `PENDING` or `BLOCKED` when evidence is absent.
- Redact credentials, personal data, private tenant details, and sensitive command output from evidence.

## Verification

- Required artifacts, metadata, IDs, and cross-references are internally consistent.
- Every active requirement maps to acceptance, implementation, and verification evidence or an explicit blocker.
- Artifact-only validation is not reported as implementation success.

## Do / Do Not

| Do | Do not |
| --- | --- |
| Preserve stable IDs, explicit source classes, and bidirectional traceability. | Reuse retired IDs or present plans as executed evidence. |
| Keep unknown status and missing evidence visible. | Invent metrics, live state, acceptance, or verification results. |

## Checklist Before Opening a PR

- [ ] The change matches this instruction's `applyTo` scope.
- [ ] Requirement, acceptance, decision, task, test, and evidence IDs resolve.
- [ ] Status and blockers match available evidence.
- [ ] Sensitive data is redacted from traceability and verification artifacts.
- [ ] Artifact checks pass without claiming implementation success.

Attribution

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

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.