create-context
/create-context
Create baseline context from .in/ folder with manifest-driven organization (run once per project). Use when bootstrapping project context, setting up .ctx/ snapshot. Triggers include "create context", "bootstrap context", "setup context", "init context".
- Author
- digital-stoic-org ↗
- Repository
- agent-skills ↗
- Licence
- MIT
- Last repository change
- 27.10.2026
Assessment and security
- Files in this package
- 4
- Scripts
- 2
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 27.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
- 2 scripts in this package. Review them before installing, as Claude may execute them.
Scripts: scripts/scan-in-folder.sh, scripts/validate-manifest.sh
What it does and when to use it
Create baseline context from .in/ folder with manifest-driven organization (run once per project). Use when bootstrapping project context, setting up .ctx/ snapshot. Triggers include "create context", "bootstrap context", "setup context", "init context".
Triggers from the description
Arguments
/create-context [--force to overwrite existing .ctx/]
Original author description
Create baseline context from .in/ folder with manifest-driven organization (run once per project). Use when bootstrapping project context, setting up .ctx/ snapshot. Triggers include "create context", "bootstrap context", "setup context", "init context".
How it works
- ⚠️ AskUserQuestion Guard
- Folder Architecture
- Steps
SKILL.md preview
---
name: create-context
description: Create baseline context from .in/ folder with manifest-driven organization (run once per project). Use when bootstrapping project context, setting up .ctx/ snapshot. Triggers include "create context", "bootstrap context", "setup context", "init context".
argument-hint: "[--force to overwrite existing .ctx/]"
allowed-tools: [Read, Write, Glob, Bash, AskUserQuestion, Task]
context: main
user-invocable: true
---
# Create Context
Bootstrap project context from `.in/` folder → `.ctx/` snapshot + baseline + RISEN INPUT table.
## ⚠️ AskUserQuestion Guard
**CRITICAL**: After EVERY `AskUserQuestion` call, check if answers are empty/blank. Known Claude Code bug: outside Plan Mode, AskUserQuestion silently returns empty answers without showing UI.
**If answers are empty**: DO NOT proceed with assumptions. Instead:
1. Output: "⚠️ Questions didn't display (known Claude Code bug outside Plan Mode)."
2. Present the options as a **numbered text list** and ask user to reply with their choice number.
3. WAIT for user reply before continuing.
## Folder Architecture
- **`.in/`**: Immutable bootstrap (user dumps raw docs, never modified by commands)
- **`.ctx/`**: Actionable snapshot (generated: manifest + summaries + copied files)
## Steps
### 1. Prerequisites
```bash
[ -d .in/ ] || error "No .in/ folder found. Create it and add source files first."
[ -d .ctx/ ] && [ "$1" != "--force" ] && error ".ctx/ exists. Use --force to recreate."
```
Skip security-sensitive files: `.env*`, `*credentials*`, `*secrets*`, `*token*`, `*.key`, `*.pem`, `*.crt`
### 2. Scan & Prioritize
1. Glob: `.in/**/*.{md,txt,csv,yaml,json}`
2. AskUserQuestion per file: HIGH / MEDIUM / LOW + brief description
### 3. Create .ctx/ and Generate Manifest
Write `.ctx/manifest.yaml` — see `reference.md` for schema.
### 4. Context Sizing & Copy
Token estimation: `tokens ≈ words / 0.75` (via `wc -w`)
| Priority | ≤ threshold | > threshold, ≤25K | > 25K |
|----------|-------------|-------------------|-------|
| HIGH | ≤1500 → inline | summarize directly | summarize via sub-agent |
| MEDIUM | ≤2500 → inline | summarize directly | summarize via sub-agent |
| LOW | reference only | — | — |
Copy HIGH/MEDIUM files to `.ctx/`, preserving subdirectory structure.
### 5. Summarize (if needed)
- Direct: Read + generate ~500 token summary → `.ctx/{nn}-{basename}-summary-llm.md`
- Sub-agent: Task(`summarize-for-context`) for files >25K tokens
### 6. Write Baseline
Create `CONTEXT-baseline-llm.md` with inline content + summary refs + LOW references.
Target: ≤2000 tokens.
### 7. Output
Report: file counts, RISEN INPUT table, suggest `/save-context baseline`.
See `reference.md` for manifest schema, baseline template, and validation rules.
Files in this package (4)
- SKILL.md 3 KB
- reference.md 4 KB
- scripts/scan-in-folder.sh 2 KB
- scripts/validate-manifest.sh 4 KB
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · digital-stoic-org/agent-skills | 100 | MIT |
| pantheon-org/tekhne | 100 | MIT |
| ducky8x/session-kit | 86 | MIT |
Database snapshot: 13.11.2026 · package version 3293ef1a62ce (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100ctx-init
Initialize AGENTS.md + CLAUDE.md + rules/ in a code repo (fresh mode). Triggers on "/ctx-init", "init context", "scaffold AGENTS.md", or when user asks to set up project context in a new repo. - 100ctx-remember
Recall project context and present structured readback. Use when the user asks 'do you remember?', at session start, or when context seems lost. - 100ctx-prompt
Apply, list, and manage saved prompt templates from .context/prompts/. Use when the user asks to apply, list, or create a reusable template like code-review or refactor. - 100project-init
[Utilities] Use when setting up or re-evaluating portable project context end to end: project-config, reference docs, CLAUDE.md, AGENTS.md. One piece only: project-config, docs-manager --mode=init or ai-context-refresh. - 100project-init
[Utilities] Use when setting up or re-evaluating portable project context end to end: project-config, reference docs, CLAUDE.md, AGENTS.md. One piece only: project-config, docs-manager --mode=init or ai-context-refresh. - 96create
Create a fresh NeatContext context from a user-defined behavioral profile and an existing local knowledge folder. Use only when the user explicitly invokes this skill or asks to create a new context rather than save the current conversation.