PiJ GitHub avatar by tonyzdev

tonyzdev /

Codebase & Graph Pathfinding

PiJ

A Pi-based terminal coding Agent whose main model handles reasoning, edits and tools while Jev provides advice.

License: MITLanguage: TypeScriptLast push: 9/19/2026#coding-agents#typed-decisions

What Jev Does in this Architecture

Jev suggests skills, reranks actual source candidates and triages failures without automatic retries or permission approval.

Key Architecture Benefit

Preserves paths, line numbers, source and errors; the author’s bounded experiments do not establish universal gains.

Review & Benchmark Note: The author provides bounded task experiments; this site reviewed implementation and evidence limits without independently retesting resolve rate, cost or latency.
Clone Repository
git clone https://github.com/tonyzdev/PiJ.git

Inspect on GitHub

Check out the upstream repository README, issues, and commit log.

tonyzdev/PiJ

Architecture Category

Codebase & Graph Pathfinding

AST traversal, code graph dependency analysis, automated PR review, and pathfinding.

View all in this category →

Related Projects in Codebase & Graph Pathfinding

Jev Review repository icon by devagrawal09

devagrawal09

Jev Review

284
Codebase & Graph PathfindingMITTypeScript

Reviews a Git diff or a codebase in stages and displays review leads in a local dashboard.

What Jev does here

Judges risk, files, evidence regions, mechanisms and severity before rule-based reviewer routing.

Connects review leads to concrete code for human inspection.

#coding-agents#ast-code-graph#evaluation-benchmarks
142 evidence links
Inspect details
Blink repository icon by ellipsis-dev

ellipsis-dev

Blink

18
Codebase & Graph PathfindingMITTypeScript

Finds files from a natural-language query using multiple walkers through the directory tree.

What Jev does here

Jev scores file and directory names; code distributes walkers using those probabilities.

Searches without a vector index and reports the share of walkers reaching each path.

#ast-code-graph#search-retrieval
3 evidence links
Inspect details
21
Codebase & Graph PathfindingMITJupyter Notebook

Navigates a Neo4j graph one hop at a time, asking Jev which relationship to follow next.

What Jev does here

Choice scores neighboring relationships and Noul checks goal completion; local beam search retains candidate paths.

Connects a natural-language goal to inspectable graph paths.

#database-vector#ast-code-graph
52 evidence links
Inspect details