> For the complete documentation index, see [llms.txt](https://hawkfi.gitbook.io/whitepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://hawkfi.gitbook.io/whitepaper/hawkfi-agents/screener-agents-osprey-v1.md). # Screener Agent (Osprey V2) Most token opportunities are easy to miss manually. Tokens move fast, rotate fast, and often become obvious only after the best entry window is gone. HawkFi Screener Agent brings 24/7 screening into HawkFi Agents, using Osprey to catch runners early and execute battle-tested models. ## What is the HawkFi Screener Agent? The Screener Agent is a HawkFi Agent type that brings 24/7 screening using Osprey to catch runners early and execute battle-tested quant models. {% hint style="success" %} The HawkFi Screener Agent's core is **Osprey** {% endhint %} ## Quick Start Tutorial {% hint style="info" %} As of June 29, 2026, `Osprey V2 High Volume 5M (Selective)` catches \~**70% of daily runners** in the current beta benchmark.\ \ Agent improvements are expected over time. {% endhint %} {% embed url="" %} Use this simplified walkthrough to create a Screener Agent with Osprey selected in the Entry Model field. {% stepper %} {% step %} #### Fund the Agent wallet. Add SOL to your `Agent funds` before turning the agent live. {% endstep %} {% step %} #### Configure Entry Settings. Set the agent name, fixed position size, maximum open positions and pools, re-entry cooldown, pool selection criteria, and optional extra filters. {% endstep %} {% step %} #### Pick the Screening agent (Osprey V2) Choose `Osprey V2 High Volume 5M (Selective)` as the Screening Agent. Osprey V2 is designed to catch runners early by screening top pools 24/7 for opportunities most humans would miss. {% endstep %} {% step %} #### Pick your Execution Model. Choose how HawkFi should execute once Osprey finds an opportunity.\ \ HawkFi screening agents use HawkFi's models as Execution Models Want to dive deeper on Screener Agent Execution Models? See [HawkFi Models](/whitepaper/models-and-automations-library/hawkfi-models.md). {% endstep %} {% step %} #### Choose Position Management settings. Add optional HawkFi automations such as Auto-Compound, Auto-Accumulate, Auto-Rebalance, or Reshape Liquidity if the position should keep managing itself after entry.\ \ Position Managements uses HawkFi's Automations.\ \ Wanna learn more about HawkFi automations? See [Automations here.](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) {% endstep %} {% step %} #### Set your Exit Settings (TP and SL) Decide whether you want to exit positions manually, use basic take profit and stop loss targets, or create custom advanced exit logic.\ \ Exit Settings use HawkFi's automations.\ \ Wanna learn more about HawkFi automations? See [Automations here.](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) {% endstep %} {% step %} #### Review and launch. Confirm the agent wallet funding, Entry Settings, Entry Model, Execution Model, position management, and exit rules before activating the agent. {% endstep %} {% endstepper %} ## What is Osprey? Osprey is the screening agent inside HawkFi Screener Agent, designed to catch runners early by screening top pools 24/7 for opportunities most humans would miss. It stays awake around the clock, watches token activity, catches daily runners, and routes winning opportunities into HawkFi agents. ### The Osprey V2 Agents
AgentBenchmarkAvg Entries / DayAvg. Losing StreakOptimized ForBehaviorStatus
Osprey High Volume 5M V17d: Catches ~67% of daily runners

14d: Catches 52% based on latest 14 days study		~10/day2Runner detectionScreens high activity pools and only enters when agent sees a qualified runner setupDeprecated
Osprey V2 High Volume 5M (Selective)14d: Catches ~70% of daily runners~3/day1.20High confidence opportunitiesPrioritizes stronger setups with fewer, more focused entries.Live beta
Osprey V2 High Volume 5M (Degen)14d: Catches ~62% of daily runners~6/day1.67Broader market opportunity coveragePrioritizes maximum opportunity coverage with faster, less selective entries.Live beta
{% hint style="info" %} **Benchmark Methodology**\ \ Benchmark results, as of June 29, 2026, are evaluated using a comprehensive 14-day historical study incorporating the latest agent improvements. Previous Osprey High Volume 5M V1 results (67% Runner Capture) were evaluated using an earlier 7-day benchmark. {% endhint %} #### What counts as a runner A runner is a top opportunity that Osprey was expected to catch. In this context, a top opportunity means a token represented by one of the highest-ranking HawkFi pools in the evaluation snapshot. These pools are usually ranked by recent pool fees, such as the latest 24-hour fees. The benchmark maps each pool to its subject token, excludes denominator assets where appropriate, and checks whether the filtered high-volume agent alerted that token inside the report window. In simple terms: ``` A runner is a top-pool token Osprey should have caught. ``` #### Current HawkFi Benchmarking Study Osprey currently catches 70% of daily runners in the beta benchmark. Osprey is not being framed around how many alerts it sends. It is being framed around how many runners it catches before they become obvious. ``` Runners caught = daily runners caught by Osprey / all eligible daily runners ``` The simple read: Osprey finds the market while humans are still watching scattered token feeds, pool pages, and noise. {% hint style="info" %} The 70% figure is the current beta benchmark for runners caught with `Osprey V2 High Volume 5M (Selective)`\ \ Agent improvements are expected over time. {% endhint %} ## Why use Osprey Most token opportunities are easy to miss manually. Tokens move fast, rotate fast, and often become obvious only after the best window is gone. HawkFi made the Osprey agent to built that gap. It monitors the market continuously 24/7, applies a repeatable screening, and catches a meaningful share of the daily runner universe before manual monitoring would usually react. {% stepper %} {% step %} ### Easy & Instant No manual setup, coding, or stitching tools together. Osprey is live inside HawkFi agents and can be configured in clicks. {% endstep %} {% step %} ### Quant Optimized No guesswork. Osprey is benchmarked against runners caught, win rate, and PnL outperformance studies, with an agent that can keep learning and improving over time. {% endstep %} {% step %} ### Battle Tested Execution Osprey routes opportunities into HawkFi's high frequency LP execution stack, built on one of the most proven LP automation engines on Solana.. {% endstep %} {% endstepper %} ## Screener Agent Building Blocks | Setup step | What it controls | | ------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Entry Settings | Choose the agent name, fixed position size, maximum open positions, [**screening agent (Osprey V2)**](#what-is-osprey), re-entry cooldown, pool selection criteria, optional extra filters, and [execution models](/whitepaper/hawkfi-agents/market-making-agents-mma/execution-models.md) | | Initial Deposit | Choose the starting liquidity shape HawkFi should use after Osprey finds a matching opportunity. | | Position Management | Add optional [automations](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) such as Auto-Compound, Auto-Accumulate, Auto-Rebalance, or Reshape Liquidity. | | Exit Settings | Choose exit [automations,](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) basic TP/SL, or custom advanced exit logic. | ### Optional extra filters for Entry Settings | Filter | What it controls | | --------------------- | ------------------------------------------------ | | Market cap | Filters opportunities by token market cap. | | Token liquidity | Filters opportunities by token-level liquidity. | | Age | Filters opportunities by token age. | | Jupiter Organic Score | Filters opportunities by Jupiter Organic Score. | | 5m volume | Filters opportunities by recent 5-minute volume. | | Fees | Filters opportunities by fee activity. | | Pool TVL | Filters opportunities by pool-level TVL. | | Base fee | Filters opportunities by pool base fee. | | Bin step | Filters opportunities by pool bin step. | {% hint style="success" icon="circle-info" %} Screener Agent uses HawkFi models as its Execution Models\ \ Learn more about [Execution models for Screener Agent here](/whitepaper/models-and-automations-library/hawkfi-models.md) {% endhint %} {% hint style="warning" icon="circle-info" %} Position Management and Exit Settings use HawkFi Automations\ \ Learn more about [HawkFi automations here](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) {% endhint %} ## FAQs for Screener Agent
I deployed the Screener Agent. Why has it not opened any position? `Osprey V2 High Volume 5M (Selective)` is selective by design. It waits for higher confidence entries instead of forcing constant positions. If you want broader opportunity coverage, `Osprey V2 High Volume 5M (Degen)` targets more entries. Higher frequency comes with tradeoffs: more entries, more activity, and higher risk.
## Related Pages * [HawkFi Agents](/whitepaper/hawkfi-agents/hawkfi-agents.md) * [Choose Your HawkFi Agent](/whitepaper/start-here/choose-your-hawkfi-agent.md) * [Fund Your Agent Wallet](/whitepaper/start-here/fund-your-agent-wallet.md) * [Screener Agent Position Management Automations](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) * [Screener Agent Exit Settings Automations](/whitepaper/models-and-automations-library/hawkfi-models/build-your-own-models.md) * [Monitor and Manage Your Agent](/whitepaper/hawkfi-agents/monitor-and-manage-your-agent.md) ## More Questions? * Join our Discord for questions and discussions on HawkFi: