# Prompt Library

> Use to save and reuse good prompts across projects and agents. Many prompts recur (a detailed feature spec, a "design the whole thing" brief, a manual/onboarding prompt) and are worth reusing verbatim or as a reference. This curates them as browsable, greppable Markdown — with a PRIVACY GATE that refuses to store anything still containing paths/emails/tokens/usernames/codenames, so the library st…

- **Type:** Skill
- **Install:** `agentstack add skill-jajupmochi-agent-harness-prompt-library`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [jajupmochi](https://agentstack.voostack.com/s/jajupmochi)
- **Installs:** 0
- **Category:** [Content & Media](https://agentstack.voostack.com/c/content-and-media)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [jajupmochi](https://github.com/jajupmochi)
- **Source:** https://github.com/jajupmochi/agent-harness/tree/main/skills/prompt-library

## Install

```sh
agentstack add skill-jajupmochi-agent-harness-prompt-library
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# prompt-library

Overhaul task 9. Reusable prompts (e.g. a very detailed feature-update brief; a "design the whole X" prompt
that transfers across sibling projects; a manual/onboarding prompt) get curated once, de-privacy'd, tagged by
scenario, and made easy for a human to find and for an agent to reuse.

## Use

- **Save** a prompt (body on stdin) — the privacy gate runs first and REFUSES if it still contains private
  content, so you de-privacy before it lands in a publishable library:
  ```
  python3 scripts/plib.py add --title "Redesign a dashboard" --scenarios ui,redesign --source claude-code /prompts/.md` (frontmatter: title/scenarios/source/tags + the de-privacy'd body) + `INDEX.md`.
Default root `recommendations/prompt-library` so the library ships with agent-harness and is reusable
cross-project. LLM-as-component: store/index/find/scan are deterministic; the model writes and adapts the
prompt text.

Status: v0.1 (add-with-privacy-gate / index / find / scan), tested (`test_plib.py`, 6/6). Planned: the first
curated batch mined from Claude Code / Codex / Copilot / opencode history (task 9 asks for this — it is
analysis work, not yet done), and per-prompt "optimized variant + when-to-use" notes.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [jajupmochi](https://github.com/jajupmochi)
- **Source:** [jajupmochi/agent-harness](https://github.com/jajupmochi/agent-harness)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-jajupmochi-agent-harness-prompt-library
- Seller: https://agentstack.voostack.com/s/jajupmochi
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.