bilalimamoglu/sift
Turn noisy command output into short, actionable diagnoses for AI coding agents.
Applies local heuristics to parse and group failures from test runners (pytest, vitest, jest), type checkers (tsc), linters (ESLint), and build tools before routing to an LLM—keeping most common cases deterministic and offline. Reduces noisy logs by 62% token usage and 71% tool calls in benchmarked debug loops by surfacing grouped root causes and anchors instead of raw output. Integrates as a CLI wrapper (`sift exec`), pipe consumer, or agent-native skill for Codex, Claude, and Cursor with optional fallback to OpenAI/OpenRouter endpoints for ambiguous cases.
Available on npm.
Stars
10
Forks
—
Language
TypeScript
License
MIT
Category
Last pushed
Mar 18, 2026
Commits (30d)
0
Dependencies
5
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/bilalimamoglu/sift"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
judge0/judge0
Robust, fast, scalable, and sandboxed open-source online code execution system for humans and AI.
kodustech/kodus-ai
AI Code Review with Full Control Over Model Choice and Costs.
web3infra-foundation/mega
Mega is an open-source implementation of Google Piper — a Git-compatible monorepo engine built...
redwoodjs/agent-ci
Agent-CI is local GitHub Actions for your agents.
probelabs/visor
Visor — AI workflow engine for code review, assistants, and automation. Orchestrates checks, MCP...