# Create Existing Solutions

> Survey existing solutions (libraries, products, internal code, standards, prior art) that address the requirements, evaluate their fit, and recommend whether to adopt, extend, or build.

- **Type:** Skill
- **Install:** `agentstack add skill-tomzx-agents-create-existing-solutions`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [tomzx](https://agentstack.voostack.com/s/tomzx)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [tomzx](https://github.com/tomzx)
- **Source:** https://github.com/tomzx/agents/tree/main/skills/create-existing-solutions

## Install

```sh
agentstack add skill-tomzx-agents-create-existing-solutions
```

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

## About

# Create Existing Solutions Survey

Surveys existing solutions that already solve part or all of what the feature needs, before committing to a design.
The goal is twofold: avoid reinventing what already exists, and harvest proven approaches as a source of information even when we end up building our own.

## Prerequisites

- Apply the shared SDLC conventions in `skills/sdlc/references/shared.md`.
- If no argument is provided, locate the feature directory under `.sdlc/features/` whose frontmatter `issue` field references `$ISSUE_NUMBER`.
- `.sdlc/features/N-/requirements.md` (must have passed review with findings verdict `approved`), or a requirements document provided in context or as a file path (`$1`)

## Steps

1. Read the requirements document and extract what the feature must do (functional requirements) and the quality attributes it must satisfy (non-functional requirements).
2. Define the search scope: the categories below, plus any project-specific sources noted in `.sdlc/context/`.
3. Search each category for candidates:
   - Existing code in this repository or organization that already does something similar (search the codebase first, it is the cheapest to adopt).
   - Open-source libraries, frameworks, and tools.
   - Commercial or SaaS products.
   - Standards, protocols, and specifications.
   - Reference material: articles, papers, and write-ups describing how others solved this problem.
4. For each candidate, capture what it is, its license, its maturity, and how it maps to the requirements (which FR/NFR it covers and which it leaves as gaps).
5. Evaluate the strongest candidates: strengths, weaknesses, integration effort, cost, and risks (license, maintenance, security, lock-in).
6. Recommend a direction (adopt, adopt and extend, build, or hybrid) with a rationale tied to the requirements.
7. Record the sources of information worth borrowing from even if we do not adopt the solution (patterns, data models, edge cases they handle, pitfalls they document).
8. Flag open questions where the evaluation is uncertain or needs validation.
9. Write the output to `.sdlc/features/N-/existing-solutions.md`.

## Output Format

Use the template at `skills/sdlc/templates/features/existing-solutions.md` (copied to `.sdlc/templates/features/existing-solutions.md` by `/initialize-sdlc-directory`; use the project's customized copy if present). Write the result to the artifact path named in the steps above.

## Outcome

If `$OUTCOME_YAML` is set, emit `verdict: approved` there per `skills/sdlc/references/shared.md`, If the artifact could not be produced, omit the file.

## Example Usage

**Scenario 1: A well-served problem**
Requirements describe rate limiting on an API.
The survey finds mature libraries (token-bucket implementations), recommends adopting one, and notes the FR it does not cover (per-tenant quotas) as a gap to build.

**Scenario 2: Reinvention avoided by internal code**
Requirements ask for CSV export.
A codebase search finds an existing export helper used elsewhere.
The recommendation is to reuse and extend it rather than write a new exporter.

**Scenario 3: Build, but learn from prior art**
Requirements describe a domain-specific scheduler with no off-the-shelf fit.
The survey recommends building, and records the data model and failure modes documented by two open-source schedulers as sources of information for the specification.

## Completion Checklist

Before handing off to review, confirm:

- [ ] Each candidate captured with license, maturity, and requirement coverage (FR/NFR) plus gaps
- [ ] Adopt/extend/build recommendation tied to the requirements

Self-check the draft against the [`review-existing-solutions` checklist](../review-existing-solutions/SKILL.md) and fix what you can, so review finds less to flag.

## Next Step

Run `/review-existing-solutions` to audit the survey for coverage, evaluation rigor, and a sound recommendation before moving on.
Once approved, continue with `/create-codebase-analysis`, which analyzes the internal code and architecture the feature will touch.

## Useful Commands Reference

| Command | Description |
|---|---|
| `WebSearch` | Find open-source projects, products, standards, and write-ups |
| `WebFetch` | Read a candidate's docs, README, license, or release history |
| `grep` / codebase search | Find existing internal code that already solves part of the problem |
| `gh search repos ...` | Search the organization's repositories for prior art (cached) |

## Source & license

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

- **Author:** [tomzx](https://github.com/tomzx)
- **Source:** [tomzx/agents](https://github.com/tomzx/agents)
- **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-tomzx-agents-create-existing-solutions
- Seller: https://agentstack.voostack.com/s/tomzx
- 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%.