required_canon_version: >=3.0.0

Skill: research-cache

Version: 0.1.0

Status: Active

Purpose

Provides persistent caching for research results to avoid redundant web browsing and API calls. Allows looking up, saving, listing, and clearing cached research summaries with optional tagging.

Trigger

Use when:

• Checking if a URL has been researched before
• Saving research summaries for future reference
• Listing cached research entries
• Clearing the research cache

Inputs

JSON object:

• action (string, required): One of: "lookup", "save", "list", "clear"

For "lookup" action:

• url (string, required): URL to look up

For "save" action:

• url (string, required): URL to save
• summary (string, required): Research summary content
• tags (string, optional): Comma-separated tags

For "list" action:

• filter (string, optional): Tag filter to narrow results

For "clear" action:

• No additional parameters

Outputs

For "lookup":

• found (boolean): Whether the URL was found in cache
• summary (string, optional): Cached summary if found
• tags (array, optional): Tags associated with entry

For "save":

• saved (boolean): Whether save was successful
• url (string): URL that was saved

For "list":

• entries (array): List of cached entries with url, summary preview, tags

For "clear":

• cleared (boolean): Whether cache was cleared

Constraints

• Governed tool: requires write access for save/clear operations
• Uses CAPABILITY/TOOLS/research_cache.py as backend
• Cache persists across sessions

Implementation

Wrapper around CAPABILITY/TOOLS/research_cache.py that:

1. Parses action and parameters
2. Calls appropriate research_cache.py subcommand
3. Returns structured JSON result

Fixtures

• fixtures/basic/ - Test list action (deterministic)

required_canon_version: >=3.0.0