Agent skill

Gsd Project Researcher

by allgpt-co in allgpt-co/QuickVoice

Researches domain ecosystem for project initialization. An agent skill from allgpt-co/QuickVoice.

MITAuto-check passedAgent Workflows

Install Gsd Project Researcher

skills CLI
$ npx skills add allgpt-co/QuickVoice --skill gsd-project-researcher -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install allgpt-co/QuickVoice gsd-project-researcher --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/allgpt-co/QuickVoice.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/gsd/agents/project-researcher .claude/skills/gsd-project-researcher && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
gsd-project-researcher
GitHub stars
488
Token cost
~2.8k tokens
SKILL.md length
766 words
Files
1
Skills in repo
12
Repo updated
First seen
Licence
MIT

At a glance

Researches domain ecosystem for project initialization. An agent skill from allgpt-co/QuickVoice.

  • Works in 5 steps: Parse Research Prompt → Determine Research Scope → Conduct Research → …
  • Tasks that involve Subagents
  • SKILL.md covers When to Use, Core Responsibilities, Philosophy and Research Dimensions, plus 6 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Gsd Project Researcher is an agent skill from allgpt-co/QuickVoice. Researches domain ecosystem for project initialization. Spawned by /gsd:new-project orchestrator (4 parallel agents).

Its SKILL.md is about 2.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows, covering Subagents. The repository describes itself as: Open-source, self-hostable platform for building and operating AI phone agents. The licence is MIT.

When your agent uses it

  • Tasks that involve Subagents

Example prompts

  • “Use the gsd-project-researcher skill to research domain ecosystem for project initialization. An agent skill from allgpt-co/QuickVoice”
  • “/gsd-project-researcher”

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Parse Research Prompt
  2. Determine Research Scope
  3. Conduct Research
  4. Write Research Document
  5. Return Confirmation

What it can do on your machine

Read from SKILL.md and the folder at commit 89fa8ff. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Gsd Project Researcher loads about 2.8k tokens when it runs. Until then it costs about 35 tokens; SKILL.md has 766 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~35
When it runs · the whole SKILL.md, loaded when a task matches
~2.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from allgpt-co/QuickVoice at commit 89fa8ff, republished under its MIT licence (© allgpt-co). 766 words, ~2,775 tokens.

Download SKILL.mdSave it as .claude/skills/gsd-project-researcher/SKILL.md (or your agent's skills folder).
name
gsd-project-researcher
description
Researches domain ecosystem for project initialization. Spawned by /gsd:new-project orchestrator (4 parallel agents).
version
1.0.0
author
GSD Project
tags
research, domain-knowledge, ecosystem-analysis
triggers
new project, research domain, discover stack
tools
Read, Write, Bash, WebFetch, mcp__context7__*

GSD Project Researcher

Researches domain ecosystem to discover standard stacks, expected features, architecture patterns, and common pitfalls.

When to Use

Use this agent when:

  • Initializing a new project with /gsd:new-project
  • Domain research is needed before defining requirements
  • You are spawned as one of 4 parallel researchers (stack, features, architecture, pitfalls)
  • Research context indicates whether this is greenfield (building from scratch) or subsequent (adding to existing app)

Core Responsibilities

  1. Research domain ecosystem - Discover what's standard for this type of project
  2. Identify table stakes - Must-have features vs differentiators vs anti-features
  3. Document architecture patterns - How systems of this type are typically structured
  4. Identify common pitfalls - What projects commonly get wrong
  5. Provide recommendations - Specific libraries, frameworks, and approaches
  6. Write research document - Create structured output for downstream consumption

Philosophy

Greenfield vs Subsequent Milestone

Greenfield (building from scratch):

  • Research: What's the standard 2025 stack for building [domain] from scratch?
  • Features: What features do [domain] products typically have?
  • Architecture: How are [domain] systems typically structured?

Subsequent (adding to existing app):

  • Research: What's needed to add [target features] to existing [domain] system?
  • Features: How do [target features] typically work?
  • Architecture: How do they integrate with existing system?
  • Don't re-research: What's already built and working
Research Quality

Be specific:

  • Recommend exact libraries with versions
  • Explain WHY each choice is recommended
  • Note compatibility requirements
  • Identify trade-offs

Be prescriptive:

  • "Use X, not Y" is better than "Consider X or Y"
  • Provide clear rationale for each recommendation
  • Include what NOT to use and why

Research Dimensions

You are spawned with a specific research type. Each researcher covers one dimension:

Stack Researcher

Research type: stack

Focus:

  • Recommended technologies and frameworks
  • Version requirements
  • Rationale for each choice
  • What NOT to use and why

Output: STACK.md

Features Researcher

Research type: features

Focus:

  • Table stakes (must-have features)
  • Differentiators (competitive advantages)
  • Anti-features (things to deliberately NOT build)
  • Feature complexity notes
  • Dependencies between features

Output: FEATURES.md

Architecture Researcher

Research type: architecture

Focus:

  • Major components and their responsibilities
  • Data flow patterns
  • Suggested build order
  • Component boundaries and communication

Output: ARCHITECTURE.md

Pitfalls Researcher

Research type: pitfalls

Focus:

  • Top 3-5 critical mistakes
  • Warning signs (how to detect early)
  • Prevention strategies (how to avoid)
  • Which phases should address each pitfall

Output: PITFALLS.md

Process

Step 1: Parse Research Prompt

Extract from your prompt:

markdown
<research_type>
Project Research — Stack dimension for [domain].
</research_type>

<milestone_context>
[greenfield OR subsequent]

Greenfield: Research standard stack for building [domain] from scratch.
Subsequent: Research what's needed to add [target features] to existing [domain] system.
</milestone_context>

<question>
What's the standard 2025 stack for [domain]?
</question>

<project_context>
[PROJECT.md summary - core value, constraints, what they're building]
</project_context>

<downstream_consumer>
Your STACK.md feeds into roadmap creation. Be prescriptive:
- Specific libraries with versions
- Clear rationale for each choice
- What NOT to use and why
Step 2: Determine Research Scope

Based on milestone context:

Greenfield:

  • Research entire ecosystem from scratch
  • Include all major technology areas
  • Provide multiple options with trade-offs

Subsequent:

  • Focus ONLY on what's needed for target features
  • Don't re-research existing system
  • Research integration patterns with existing codebase
Show full SKILL.md (346 more words)Show less
Step 3: Conduct Research

Use Context7 MCP or WebSearch to research:

For Stack Researcher:

  • Search for: "[domain] 2025 stack", "[domain] framework recommendations"
  • Check official documentation for latest versions
  • Compare multiple options
  • Identify version constraints and compatibility

For Features Researcher:

  • Search for: "[domain] SaaS features", "[domain] application features"
  • Analyze competitor products
  • Identify table stakes vs differentiators
  • Research feature complexity and implementation effort

For Architecture Researcher:

  • Search for: "[domain] architecture patterns", "[domain] system design"
  • Study reference implementations and case studies
  • Identify common architectural approaches
  • Document best practices for this domain

For Pitfalls Researcher:

  • Search for: "[domain] common mistakes", "[domain] pitfalls", "building [domain] errors"
  • Research typical failure modes
  • Identify early warning signs
  • Document prevention strategies
Step 4: Write Research Document

Use template: ./.claude/get-shit-done/templates/research-project/[DIMENSION].md

Document structure:

  • Executive summary (2-3 paragraphs)
  • Key findings organized by category
  • Confidence levels for each recommendation
  • Specific, actionable recommendations
Step 5: Return Confirmation

Return brief confirmation:

markdown
## RESEARCH COMPLETE

**Research Type:** [stack | features | architecture | pitfalls]
**Output:** .planning/research/[DIMENSION].md

**Key Findings:**
- [Finding 1]
- [Finding 2]
- [Finding 3]

Ready for synthesizer.

Document Templates

STACK.md Template
markdown
# Technology Stack

**Analysis Date:** [YYYY-MM-DD]

## Languages

**Primary:**
- [Language] [Version] - [Where used]

**Secondary:**
- [Language] [Version] - [Where used]

## Runtime

**Environment:**
- [Runtime] [Version]

**Package Manager:**
- [Manager] [Version]
- Lockfile: [present/missing]

## Frameworks

**Core:**
- [Framework] [Version] - [Purpose]

**Testing:**
- [Framework] [Version] - [Purpose]

**Build/Dev:**
- [Tool] [Version] - [Purpose]

## Key Dependencies

**Critical:**
- [Package] [Version] - [Why it matters]

**Infrastructure:**
- [Package] [Version] - [Purpose]

## Configuration

**Environment:**
- [How configured]
- [Key configs required]

**Build:**
- [Build config files]

## Platform Requirements

**Development:**
- [Requirements]

**Production:**
- [Deployment target]

---

*Stack analysis: [date]*
FEATURES.md Template
markdown
# Features Analysis

**Analysis Date:** [YYYY-MM-DD]

## Table Stakes

**Must-have features** (users expect these or they leave):
- [Feature 1] - [Brief description]
- [Feature 2] - [Brief description]
- [Feature 3] - [Brief description]

**Differentiators** (competitive advantages):
- [Feature 1] - [Why this gives you an edge]
- [Feature 2] - [Why this is valuable]
- [Feature 3] - [Brief description]

**Anti-features** (things to deliberately NOT build):
- [Feature 1] - [Why not to build this]
- [Feature 2] - [Why defer or skip]

## Feature Complexity

[Notes on implementation effort and dependencies]

---

*Features analysis: [date]*
ARCHITECTURE.md Template
markdown
# Architecture Analysis

**Analysis Date:** [YYYY-MM-DD]

## Pattern Overview

**Overall:** [Pattern name]

**Key Characteristics:**
- [Characteristic 1]
- [Characteristic 2]
- [Characteristic 3]

## Layers

**[Layer Name]:**
- Purpose: [What this layer does]
- Location: `[path]`
- Contains: [Types of code]
- Depends on: [What it uses]
- Used by: [What uses it]

## Data Flow

**[Flow Name]:**

1. [Step 1]
2. [Step 2]
3. [Step 3]

**State Management:**
- [How state is handled]

## Key Abstractions

**[Abstraction Name]:**
- Purpose: [What it represents]
- Examples: `[file paths]`
- Pattern: [Pattern used]

## Entry Points

**[Entry Point]:**
- Location: `[path]`
- Triggers: [What invokes it]
- Responsibilities: [What it does]

## Error Handling

**Strategy:** [Approach]

**Patterns:**
- [Pattern 1]
- [Pattern 2]

## Cross-Cutting Concerns

**Logging:** [Approach]
**Validation:** [Approach]
**Authentication:** [Approach]

---

*Architecture analysis: [date]*
PITFALLS.md Template
markdown
# Common Pitfalls

**Analysis Date:** [YYYY-MM-DD]

## Critical Pitfalls

**1. [Pitfall Name]**
   - **Warning Signs:** [How to detect early]
   - **Prevention Strategy:** [How to avoid]
   - **Which Phase Should Address:** [Phase number]

**2. [Pitfall Name]**
   - **Warning Signs:** [How to detect early]
   - **Prevention Strategy:** [How to avoid]
   - **Which Phase Should Address:** [Phase number]

**3. [Pitfall Name]**
   - **Warning Signs:** [How to detect early]
   - **Prevention Strategy:** [How to avoid]
   - **Which Phase Should Address:** [Phase number]

---

*Pitfalls analysis: [date]*

Quality Gates

Before returning research complete, ensure:

  • Versions are current (verify with Context7/official docs, not training data)
  • Rationale explains WHY, not just WHAT
  • Confidence levels assigned to each recommendation
  • Specific libraries with versions recommended
  • Clear trade-offs identified
  • What NOT to use is documented with reasons
  • Research document follows template structure
  • All findings are actionable and specific

Critical Rules

  • Focus on research type - Stick to your dimension (stack, features, architecture, or pitfalls)
  • Be prescriptive - Recommend specific approaches, not list options
  • Consider milestone context - Greenfield vs subsequent research scope differs
  • Use Context7 - For library/API documentation and version verification
  • Document trade-offs - Explain why certain choices are recommended
  • Be specific - Provide exact library names and versions when possible
  • Follow template - Use the provided template structure

Success Criteria

  • Research prompt parsed correctly
  • Research scope determined based on milestone context
  • Domain research conducted using Context7/WebSearch
  • Key findings identified and organized
  • Recommendations are specific and actionable
  • Confidence levels assessed honestly
  • Research document written to correct location
  • Document follows template structure
  • Confirmation returned (not document contents)
  • @skills/gsd/agents/research-synthesizer - Agent that synthesizes your output
  • @skills/gsd/agents/roadmapper - Agent that uses your research to create roadmap
  • @skills/gsd/commands/new-project - Command that spawns you

© allgpt-co, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/gsd/agents/project-researcher of allgpt-co/QuickVoice.

Open the folder on GitHubat commit 89fa8ff

Compare with similar skills

Gsd Project Researcher next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Gsd Project Researcher compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Gsd Project Researcher this skillallgpt-co/QuickVoice488—~2.8kAutomated safety check: PassMIT
Synthesisyologdev/yoyo-evolve1.9k—~3.1kAutomated safety check: PassMIT
Ad Subagentalexandremendoncaalvaro/CorridorKey-Runtime7561 repos~1.3kAutomated safety check: PassCustom licence
Subagent Driven Reviewwentorai/Research-Claw858—~3kAutomated safety check: PassCustom licence
Ad SubagentCorridorTech/PoseCap224—~1.3kAutomated safety check: NotesApache-2.0
RoundtableLeoYeAI/openclaw-master-skills2.2k—~5kAutomated safety check: PassMIT

Similar skills

  • Synthesis

    yologdev/yoyo-evolve

    Multi-source research synthesis — aggregate and compare 3+ sources or any source 5KB using sub-agent dispatch and SharedState

    1.9k GitHub stars~3.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Ad Subagent

    alexandremendoncaalvaro/CorridorKey-Runtime

    Draft a new Codex custom subagent at .codex/agents/<name.toml, using the official Codex subagents format.

    756 GitHub starsUsed in 1 repo~1.3k tokens
    Agent WorkflowsAuto-check passed
  • Subagent Driven Review

    wentorai/Research-Claw

    Use parallel subagents to scale large reviews, with an honest model of where the speedup actually comes from (LLM reasoning, not API throughput)

    858 GitHub stars~3k tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed
  • Ad Subagent

    CorridorTech/PoseCap

    Draft a new Claude Code subagent at .claude/agents/<name.md, using the official subagents format.

    224 GitHub stars~1.3k tokensUpdated 4 days ago
    Agent WorkflowsAuto-check: notes
  • Roundtable

    LeoYeAI/openclaw-master-skills

    Multi-agent debate council — spawns 3 specialized sub-agents in parallel (Scholar, Engineer, Muse) for Round 1, then optional Round 2 cross-examination to challenge assumptions and strengthen the…

    2.2k GitHub stars~5k tokensUpdated 2 mo ago
    Agent WorkflowsAuto-check passed
  • Claude Code Agent Development

    anthropics/claude-plugins-official

    Official

    Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.

    38k GitHub starsUsed in 7 repos~2.8k tokens
    Agent WorkflowsAuto-check passed

More from allgpt-co/QuickVoice

All 12 skills in this repo
  • Livekit Agents

    allgpt-co/QuickVoice

    Build voice AI agents with LiveKit Cloud and the Agents SDK.

    488 GitHub stars~3.3k tokensUpdated 5 days ago
    Auto-check: notes
  • Gsd

    allgpt-co/QuickVoice

    Get Shit Done (GSD) - A comprehensive project management system for solo developers using Claude agents

    488 GitHub stars~2k tokensUpdated 5 days ago
    Auto-check passed
  • Gsd Codebase Mapper

    allgpt-co/QuickVoice

    Explores codebase and writes structured analysis documents. An agent skill from allgpt-co/QuickVoice.

    488 GitHub stars~3.5k tokensUpdated 5 days ago
    Auto-check: notes
  • Gsd Integration Checker

    allgpt-co/QuickVoice

    Verifies that integrations work correctly by checking endpoints, responses, and data flow.

    488 GitHub stars~2.8k tokensUpdated 5 days ago
    Auto-check passed
  • Gsd Phase Researcher

    allgpt-co/QuickVoice

    Researches phase implementation for planning. An agent skill from allgpt-co/QuickVoice.

    488 GitHub stars~1.9k tokensUpdated 5 days ago
    Auto-check passed
  • Gsd Plan Checker

    allgpt-co/QuickVoice

    Validates plan quality by checking task completeness, dependency correctness, and scope sanity.

    488 GitHub stars~2.5k tokensUpdated 5 days ago
    Auto-check passed

Categories

Questions about Gsd Project Researcher

What does Gsd Project Researcher do?

Researches domain ecosystem for project initialization. An agent skill from allgpt-co/QuickVoice. Gsd Project Researcher is an agent skill from allgpt-co/QuickVoice. Researches domain ecosystem for project initialization.

When should I use Gsd Project Researcher?

Gsd Project Researcher fits situations like: tasks that involve Subagents.

How do I install Gsd Project Researcher in Claude Code?

Run `npx skills add allgpt-co/QuickVoice --skill gsd-project-researcher -a claude-code`. Or copy the skill folder (.claude/skills/gsd/agents/project-researcher in allgpt-co/QuickVoice) into .claude/skills/gsd-project-researcher in your project. Claude Code loads it when a task matches its description.

How do I install Gsd Project Researcher in Codex?

Run `npx skills add allgpt-co/QuickVoice --skill gsd-project-researcher -a codex`. Or copy the skill folder (.claude/skills/gsd/agents/project-researcher in allgpt-co/QuickVoice) into .agents/skills/gsd-project-researcher in your project. Codex loads it when a task matches its description.

Can I use Gsd Project Researcher in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add allgpt-co/QuickVoice --skill gsd-project-researcher -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/gsd-project-researcher, .gemini/skills/gsd-project-researcher, .github/skills/gsd-project-researcher and .opencode/skills/gsd-project-researcher in your project.

What does Gsd Project Researcher need to run?

SKILL.md names no scripts, command-line tools or credentials: Gsd Project Researcher is instructions for the agent only.

Does Gsd Project Researcher access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Gsd Project Researcher safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Gsd Project Researcher use?

Gsd Project Researcher is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Gsd Project Researcher use?

About 2.8k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Gsd Project Researcher?

Skills that share tags, products or a category with Gsd Project Researcher: Synthesis (yologdev/yoyo-evolve, 1.9k stars), Ad Subagent (alexandremendoncaalvaro/CorridorKey-Runtime, 756 stars), Subagent Driven Review (wentorai/Research-Claw, 858 stars) and Ad Subagent (CorridorTech/PoseCap, 224 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Gsd Project Researcher?

allgpt-co (a GitHub organization) maintains it in allgpt-co/QuickVoice, which has 488 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on October 5, 2026.

Source: allgpt-co/QuickVoice on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.