ml4t-agent-tool-contracts
Typed tool contracts for autonomous research agents. Use when exposing files, search, databases, or execution tools to an LLM agent.
38,472 Skills
Search uses this published catalog; rankings and counts are snapshots.
Typed tool contracts for autonomous research agents. Use when exposing files, search, databases, or execution tools to an LLM agent.
Commission, slippage, and market-impact cost models for realistic strategy simulation. Use when backtesting to ensure P&L accounts for transaction costs.
Multi-agent probability forecasting with diversity, aggregation, and debate controls. Use when combining several agent forecasts or evaluating forecast ensembles.
Thin autonomous research operator pattern for ML4T experiments. Use when an agent should inspect artifacts, read skills, run scripts, and produce an auditable recommendation.
Reinforcement learning for trade execution and hedging. Use when optimizing execution algorithms or dynamic hedging policies.
Event-driven backtesting with realistic order execution, position tracking, and performance measurement. Use when simulating a trading strategy on historical data.
Transform any research material into SEO+GEO-optimized blog posts for Knox Hub. Takes any input (URL, article text, notes, video transcript, competitor post, topic idea) and rewrites it as natural human-sounding content, then fills in all fields for the Knox Hub post editor: titl
Use when the user wants an app idea turned into a Product Requirements Document (PRD) committed to a repo — "create a PRD", "/app-prd", "turn this idea into a PRD", or a raw idea dump ending with "save it to this repo". Produces a complete, section-numbered PRD.md landed via work
Use when the user asks to "create a release", "cut a release", "tag a version", "publish release notes", or ship a new versioned GitHub release for the current repo.
Use when the user wants to review or revise the ordering/model/effort/fableplan/plan-effort assignments on a milestone's GitHub issues — "review the execution plan", "/execution-plan-review", "show me the model assignments", or piecemeal revisions like "11 should be medium" or "p
Use when the user wants a task executed by the current session model with a persistent Fable 5.1 advisor overseeing it. A read-only Fable 5.1 subagent writes the plan and answers checkpoint consults; a separate fresh Fable 5.1 reviewer issues a binding pre-commit verdict. Accepts
Required dispatch procedure for running a subagent on Fable 5.1: positive harness detection, the Claude Code CLI shim for other harnesses, the fallback ladder, the result-parsing contract, the attribution rule, and the caller dispatch-hygiene rules (read-only prompt, snapshot/dif
Use when the user wants a GitHub issue created by a Fable 5.1 subagent. Spins up a read-only subagent running on Fable 5.1 that executes the new-issue procedure (duplicate check, code grounding, approach design, complexity score) and returns a fully-composed issue draft, which th
Use when the user asks to have a Fable 5.1 subagent draft and file a GitHub issue, then autonomously drive it all the way to a reviewed PR in one shot — "fable-new-issue-loop", "fable create the issue and run it to completion". Runs fable-new-issue to create a fully-specified iss
Use when the user wants a task decomposed and driven by a Fable 5.1 orchestrator delegating implementation to Sonnet 5 workers. Fable decomposes into self-contained specs, dispatches Sonnet workers (Agents for 1–2 pieces, a Workflow for fan-out), reviews each result, merges into
Use when the user wants a GitHub issue validated by a Fable 5.1 subagent. Spins up a read-only subagent running on Fable 5.1 that executes the validate-issue procedure (claim tracing, architecture/consistency checks, complexity score), then relays the verdict back to the main age
Catalog captured:2026-11-21