AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Context Builder Cowork

skill-kirkruglov-claude-skills-kit-context-builder-cowork · by KirKruglov

Generates a ready-to-use project-context.md file for Claude Cowork through an interactive interview. Triggers EN: 'create project context', 'new project context', 'context-builder', 'generate context file', 'quick context'. Triggers RU: 'создай контекст проекта', 'новый проект контекст', 'сформируй контекстный файл', 'быстрый контекст'.

No reviews yet
0 installs
37 views
0.0% view→install

Install

$ agentstack add skill-kirkruglov-claude-skills-kit-context-builder-cowork

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-kirkruglov-claude-skills-kit-context-builder-cowork)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Context Builder Cowork? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Context Builder Cowork

Generates a ready-to-use project-context.md for Claude Cowork based on user answers through a structured interview. Supports two modes: full interview (15 questions across 9 blocks) and quick mode (first 3 blocks only, rest filled with placeholders).

For: anyone who works with Claude Cowork and needs a structured project context file for a new or existing project.

Language Detection

Detect the language of the user's first message — use it throughout the entire session: all interview questions, the generated file content, and section headers. Russian input → conduct interview and generate file in Russian. English input → conduct interview and generate file in English.

Input

Required:

  • User's message requesting context creation (sets language and mode)

Optional:

  • Keyword "quick" / "быстрый" in the trigger → activates quick mode (blocks 1–3 only, rest → [to clarify] / [уточнить])

Output

A ready-to-download project-context.md file structured across up to 9 content blocks, with [to clarify] / [уточнить] placeholders for missing information and a changelog entry with creation date.

Instructions

Step 1 — Determine mode

If the user wrote "quick context" / "быстрый контекст" → use quick mode (questions 1–8, blocks 1–3 only). Otherwise → use new mode (all 15 questions, all 9 blocks).

| Mode | Trigger | What it does | |------|---------|--------------| | new | "create project context" / "создай контекст проекта" | Full interview, all blocks | | quick | "quick context" / "быстрый контекст" | Blocks 1–3 only, rest → placeholders |

Step 2 — Conduct the interview

Ask questions in blocks, not all at once. Wait for the user's response after each block, then proceed to the next.

Block 1 — Basics
1. Project name?
2. Type: business / analytics / personal / other?
3. Status: Active / On hold / Completed?
4. Start date? Deadline (if any)?
Block 2 — Project essence
5. Describe the project in 2–4 sentences.
6. Vision — what outcome counts as project success?
Block 3 — Current stage goal
7. What are we building / working on right now?
8. What is the specific deliverable for this stage?

quick mode stops here. Remaining blocks are filled with placeholders.

Block 4 — Project structure
9. Describe the key elements, components, or layers of the project (freely, any format).
Block 5 — Participants and dependencies
10. Who is involved? List: stakeholders, partners, external dependencies — briefly for each.
Block 6 — Constraints
11. Are there any constraints? Time, resources, risks — what matters?
Block 7 — Data structure
12. Path to the input folder? List key files if any.
13. Path to the output folder? List already created files if any.
Block 8 — Open questions
14. Are there known open questions or unresolved tasks? (Any format works)
Block 9 — Additional sections (optional)
15. Do you need additional sections?
    - Market context
    - Methodology
    - Stages / Roadmap
    Or none of the above?

If the user selects additional sections — ask 1–2 clarifying questions for each selected section.

Step 3 — Generate the file

Processing rules:

  • Do not paste user answers verbatim — write each block as coherent structured text
  • Preserve the user's terminology and phrasing
  • Do not invent facts — only what the user stated
  • Missing or unclear items → mark as [to clarify] (EN) or [уточнить] (RU)
  • Changelog: one entry with the file creation date

Output filename: project-context.md

Step 4 — Deliver the file

  1. Create the file project-context.md with the generated content
  2. Make it available for download
  3. Briefly report: how many blocks are filled, how many are marked with placeholders

Output Format

# Project Context: [name]

## Basics
- **Type:** [value]
- **Status:** [value]
- **Start date:** [value]
- **Deadline:** [value or "Not defined"]

---
# Project Overview
[Coherent text based on answer 5]

# Vision
[Coherent text based on answer 6]

---
# Current Stage Goal
[Coherent text based on answers 7–8]

---
# Project Structure
[Coherent text based on answer 9]

---
# Participants and Dependencies
[Structured text based on answer 10]

---
# Constraints
[Text based on answer 11]

---
# Data Structure

Input: [input/ path]
Files: [list from answer 12]

Output: [output/ path]
Files: [list from answer 13]

---
# Open Questions
[List from answer 14, formatted as a tracker]

---
[Optional sections if selected]

---
# Changelog

| Date | Change |
|------|--------|
| [date] | Project context created |

Negative Cases

  • Do not activate when the user is asking about an existing project context, not creating a new one
  • Do not generate the file before completing the interview
  • If the user provides all information in one message — do not force an interview; assemble the file directly from the provided information

Source & license

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

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

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.