context-researchlisted

# Context Research Skill ## Purpose Eliminate the "let me look that up" cycle. Research a topic once, persist the findings, and let Claude reference them automatically in future sessions. ## Configuration Check for `.context/config.md` in the project root. If it exists, read the YAML frontmatter for source priority and preferences. For the default config template, see `references/source-config-template.md`. ## Workflow ### Step 1: Parse Input Extract from `$ARGUMENTS`: - **Topic**: The subject to research (e.g., "Stripe subscriptions with Next.js") - **Refresh flag**: Whether `--refresh` was passed ### Step 2: Load Config Read `.context/config.md` if it exists, parse YAML frontmatter for source priority. Fall back to defaults if missing. ### Step 3: Check Existing Context ```bash # Slugify the topic: lowercase, spaces→hyphens, strip special chars SLUG=$(echo "$TOPIC" | tr '[:upper:]' '[:lower:]' | sed 's/[^a-z0-9]/-/g' | sed 's/--*/-/g' | sed 's/^-//;s/-$//') ``` Check if `.context/${SLUG}.md` already exists: - If exists AND no `--refresh` → show the existing brief, note its date, ask if user wants to refresh - If exists AND `--refresh` → proceed with re-research - If not exists → proceed with research ### Step 4: Research (Using Configured Source Priority) Iterate through the `sources` list from config. For each source: **context7** (if in list and available): - `resolve-library-id` for the library/framework mentioned in topic - `query-docs` with relevant que