In Progress
v0.12 — .env secrets workaround
Inject credentials at runtime via Docker --env-file. AI agents never see user secrets.
Arun Raghunath about 1 month ago
In Progress
v0.12 — .env secrets workaround
Inject credentials at runtime via Docker --env-file. AI agents never see user secrets.
Arun Raghunath about 1 month ago
Completed
v0.10 — Developer CLI
Run any command against a live sandbox straight from your terminal. jhansi watch syncs your project, jhansi exec runs it. No uploads, no sandbox IDs, no friction.
Arun Raghunath about 1 month ago
Completed
v0.10 — Developer CLI
Run any command against a live sandbox straight from your terminal. jhansi watch syncs your project, jhansi exec runs it. No uploads, no sandbox IDs, no friction.
Arun Raghunath about 1 month ago
Completed
v0.9 — Observability & Execution History
Every execution now leaves a trace. jhansi.io gains a full run model — every code execution is stored, searchable, and never silently lost. What's in it: Typed SSE events — output and done event types, so AI agents can act on structured results not raw text Run model — every execution stored with full output, exit code, duration, and error summary Soft delete — sandboxes are retained after deletion, full history preserved Agent tracking — tag sandboxes with agent and created_by at creation time Metrics endpoint — live counts, success rate, average duration, broken down by agent Rolling log retention — configurable storage cap, oldest logs auto-evicted when limit hit
Arun Raghunath about 1 month ago
Completed
v0.9 — Observability & Execution History
Every execution now leaves a trace. jhansi.io gains a full run model — every code execution is stored, searchable, and never silently lost. What's in it: Typed SSE events — output and done event types, so AI agents can act on structured results not raw text Run model — every execution stored with full output, exit code, duration, and error summary Soft delete — sandboxes are retained after deletion, full history preserved Agent tracking — tag sandboxes with agent and created_by at creation time Metrics endpoint — live counts, success rate, average duration, broken down by agent Rolling log retention — configurable storage cap, oldest logs auto-evicted when limit hit
Arun Raghunath about 1 month ago
Completed
v0.8 — MCP Server
Model Context Protocol server for Claude Code and Cursor integration. Run AI-generated code safely inside your editor without leaving your workflow. One MCP server, works across all major AI coding tools.
Arun Raghunath about 1 month ago
Completed
v0.8 — MCP Server
Model Context Protocol server for Claude Code and Cursor integration. Run AI-generated code safely inside your editor without leaving your workflow. One MCP server, works across all major AI coding tools.
Arun Raghunath about 1 month ago
Completed
v0.7 — Streaming Execution
Real-time output streaming for sandbox execution using Server-Sent Events. Watch dependency installation and code output live instead of waiting for completion. Removes the 300s timeout workaround.
Arun Raghunath about 1 month ago
Completed
v0.7 — Streaming Execution
Real-time output streaming for sandbox execution using Server-Sent Events. Watch dependency installation and code output live instead of waiting for completion. Removes the 300s timeout workaround.
Arun Raghunath about 1 month ago
Completed
v0.6 — Persistent Registry + TTL
Petri's sandbox registry is currently in-memory — a restart wipes all state. v0.6 fixes that. What's changing: File or DB-backed registry so sandboxes survive restarts TTL per sandbox — inactive containers are torn down automatically Prevents disk and memory leaks from abandoned sandboxes Why it matters: The SDK and any long-running agent workflow needs to trust that a sandbox it created is still there. Without persistence, you can't build reliably on top of Petri.
Arun Raghunath about 1 month ago
Completed
v0.6 — Persistent Registry + TTL
Petri's sandbox registry is currently in-memory — a restart wipes all state. v0.6 fixes that. What's changing: File or DB-backed registry so sandboxes survive restarts TTL per sandbox — inactive containers are torn down automatically Prevents disk and memory leaks from abandoned sandboxes Why it matters: The SDK and any long-running agent workflow needs to trust that a sandbox it created is still there. Without persistence, you can't build reliably on top of Petri.
Arun Raghunath about 1 month ago
Completed
v0.4 — Test Mode
Upload your project as a zip, pass a command — jhansi starts it and runs your test suite against it. Works for scripts and servers alike.
Arun Raghunath about 1 month ago
Completed
v0.4 — Test Mode
Upload your project as a zip, pass a command — jhansi starts it and runs your test suite against it. Works for scripts and servers alike.
Arun Raghunath about 1 month ago