bricelancasterwcp-sudo/sensorium
by Various
Record what a program actually did and ask it questions afterward — Python, Rust and TypeScript recorders writing one trace format, read by one LLM-native query CLI that refuses ra
MCP
bricelancasterwcp-sudo/sensorium
Added 23 Sept 2026
Overview
Records what a program actually did using Python, Rust, and TypeScript recorders that write a single trace format. An LLM-native query CLI reads the trace and answers questions about it, refusing rather than guessing when it lacks evidence.
Best for
Best for
Developers debugging multi-language programs through natural language trace queries
Use cases
- Ask natural language questions about a recorded program run
- Debug past executions without adding log statements
- Analyze traces from Python, Rust, and TypeScript in one tool
Notes
0 stars on GitHub. Last updated 2026-09-20. Licensed MIT.
Pros
- Unified trace format across three languages
- Query CLI avoids hallucination by refusing uncertain answers
- Simple workflow: record, then interrogate
Cons
- Only supports Python, Rust, and TypeScript recorders
- Zero stars indicates minimal community adoption or validation
- LLM may decline to answer questions not grounded in the trace
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.