# Project Research

> Research and data-gathering phase for {PROJECT} — domain knowledge, reference data models, external API details, and a gap analysis against what already exists. Called by project-orchestrator when a module/topic's knowledge is missing; also used directly when the user says "research {PROJECT}'s X module.

- **Type:** Skill
- **Install:** `agentstack add skill-cormff-claude-code-playbook-project-research`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [cormff](https://agentstack.voostack.com/s/cormff)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [cormff](https://github.com/cormff)
- **Source:** https://github.com/cormff/claude-code-playbook/tree/main/skill-tree/layer-2-orchestrator/project-research

## Install

```sh
agentstack add skill-cormff-claude-code-playbook-project-research
```

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

## About

# {PROJECT} Research

Produces one accessible, cross-linked vault note per module/topic — not a one-off answer that disappears at the end of the conversation.

## Step 1 — Scope the module/topic

Identify what's being researched and where its note should live under `{PROJECT_NOTES_PATH}`.

## Step 2 — Gather

- Domain knowledge relevant to this module.
- Reference data models or prior art, internal or external.
- Any external system/API details this module needs to integrate with.

Delegate to a subagent whenever this involves scanning multiple sources — see [05 — Agents & Delegation](../../../docs/05-agents-and-delegation.md).

## Step 3 — Gap analysis

Explicitly compare "what exists already" against "what this module needs" and call out the gap — this is the part most likely to get skipped if research turns into a generic summary instead of decision-relevant input for the design phase.

## Step 4 — Write the note

Follow [`vault-format`](../../layer-0-foundation/vault-format/SKILL.md) conventions; link it from the project's map/hub. This note is what [`project-design`](../project-design/SKILL.md) will read next — write it for that purpose, not as a narrative transcript of the research process.

## Source & license

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

- **Author:** [cormff](https://github.com/cormff)
- **Source:** [cormff/claude-code-playbook](https://github.com/cormff/claude-code-playbook)
- **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-cormff-claude-code-playbook-project-research
- Seller: https://agentstack.voostack.com/s/cormff
- 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%.