agentfootprint
TypeScript AI agent framework with recorded context, tool calls and decisions, plus configurable answer checks.
- 22
- TypeScript
- MIT
- 7 h ago
TypeScript AI agent framework with recorded context, tool calls and decisions, plus configurable answer checks.
TypeScript workflow engine with named stages, tracked state and recorded decision evidence for execution debugging.
TypeScript action and journey graphs for agent-accessible web apps, with live availability and recorded interaction outcomes.
Execution records, replay, time travel and honest attribution — independent of the engine that produced them
Experimental Beta: coordinated visualization and analysis with selection provenance, recorded interactions and evidence inspection. Source checkout only; no npm release.
Why Lens: React components for inspecting recorded agentfootprint prompts, tool calls, context and decisions.
TypeScript commerce workflows and business profiles on Blueprint: scoped actions, shipping, after-sales and retained operation evidence. See README for source-candidate status.
TypeScript framework for configurable multi-tenant platforms: scoped profiles, feature modules, adapter routing, actions and retained execution evidence. See README for source-candidate status.
React components for inspecting FootPrint workflow execution, recorded state changes and time-travel debugging.
Beta ML evidence library for Python and TypeScript: model and dataset references, evaluation receipts, release guards and saved-workflow investigation.
Home of footprintjs: execution provenance for AI agents and software. A film, the library list and the thinking behind it.
Footprint ecosystem profile: workflow libraries, agent tooling, recorded evidence and project navigation.