Oregon Code

Several tools have heavily overlapping purposes: ask_pipeworx and ask_pipeworx_beta are currently functionally identical, and the five polymarket_* tools (arbitrage, edges, edge_tracker, fill_risk, kalshi_spread) all relate to finding or evaluating prediction-market mispricings with fuzzy boundaries. Company research is split across entity_profile, company_facts, compare_entities, and recent_changes, which could lead to misselection.

All tool names use consistent snake_case, and several follow clear prefixes (ask_*, polymarket_*, scan_*, resolution_*). However, conventions are mixed: many are verb-based (subscribe, recall, validate_claim) while others are noun phrases (entity_profile, company_facts, release_calendar_markets), and a few like or_revised_statute and pipeworx_trending don't fit any obvious pattern.

At 37 tools, the set is well beyond the 16-25 range that already feels heavy, pushing into the 'too many' territory from the calibration. The broad data-research scope justifies some bulk, but the count is bloated by redundant meta-tools and near-duplicate prediction-market helpers.

The tool surface is remarkably comprehensive: data lookup, grounded verification, company research, entity resolution, prediction-market analysis with arbitrage and settlement auditing, subscriptions lifecycle (subscribe/unsubscribe/list/alerts), memory (remember/recall/forget), and utility tools like visibility checks and dependency scanning. There are few obvious gaps given the stated scope.