Free agent finder

Skill garavitgabriel/espn-fantasy-claude-openclaw/skills/free-agent-finder

Finds and recommends free agents when the user needs a player, asks about the waiver wire, needs a position filled, wants to stream a pitcher, or asks 'who should I pick up'. Searches by position, compares top options, and recommends based on H2H category impact. Triggers on: 'free agent', 'waiver', 'pickup', 'pick up', 'stream', 'available', position needs like 'I need a shortstop'.From its SKILL.md

Install

npx -y skills add garavitgabriel/espn-fantasy-claude-openclaw --skill free-agent-finder

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.4 KB, as published. Nobody here has run it

Free Agent Finder

When the user is looking for free agent pickups or waiver wire targets, follow these steps:

Step 1: Understand the League Categories

Call get_scoring_categories to know which 14 stats matter and their direction (some are reverse — lower wins).

Step 2: Assess Current Roster

Call get_my_roster to see the current team composition, stats, and any gaps.

Step 3: Identify Needs

If the user specified a position, use that position.
If not, call get_roster_needs to identify which positions are unfilled or weakest.
Check memory if available: call get_category_trends to identify which categories need help, and get_watchlist to see if there are players already being tracked.

Step 4: Search Free Agents

Call get_free_agents with the target position filter. Request up to 25 results to have a good pool.

Step 5: Deep Evaluation

For the top 3-5 candidates from the free agent list, call get_player_info for each to get detailed stat breakdowns. If comparing two close options, use compare_players.

Step 6: Present Recommendations

Rank the candidates by category impact, not just total points. For each recommendation:

Player name and position
Key categories they help: Which of the 14 categories will improve?
Categories they might hurt: e.g., a power hitter with high B_SO (bad in a lower-wins category)
Ownership %: How likely is someone else to grab them?
Verdict: Clear recommendation on who to pick up and who to drop

For the reverse categories (B_SO, ERA, WHIP, L), remember that lower values are better. A pitcher with high K but also high ERA is a mixed bag.

Step 7: Log to Memory (if available)

If the user decides to pick someone up, call save_roster_move with action="ADD" and the player details
If they drop someone, call save_roster_move with action="DROP"
If a candidate was on the watchlist, call remove_from_watchlist after pickup
If they want to monitor someone for later, call add_to_watchlist