enricai

7 mods across 1 repository, 10 stars between them.

enricai-leerie

01

enricai/leerie

Plugin Claude Code

Plugin marketplace listing 1 plugin: leerie.

10 5d ago A tokens not measured original MIT

leerie

02

enricai/leerie

Plugin Claude Code

Graph-engineered task driver for Claude Code. Classifies a task, decomposes it into a dependency graph of granular subtasks, topo-sorts it into waves, and executes each node in an isolated git worktree under an evidence-gated implement/validate loop. Runs entirely on the Claude Code CLI/subscription via claude -p wor.

10 5d ago A tokens not measured original MIT

leerie CLAUDE.md

03

enricai/leerie

Instructions file

Instructions for enricai/leerie, covering claude.md, tech stack, the three-layer rule (load-bearing — read first), the central principle: prompts are advisory, code enforces and prompts must not name another project's code.

10 5d ago A 7,385 tokens original MIT

chain

04

enricai/leerie

Command

Submit and manage a multi-run leerie chain. Use when the user wants to run a sequence of leerie runs across N sequential waves driven by the laptop.

10 5d ago A 32 tokens original MIT

leerie

05

enricai/leerie

Command

Launch the Leerie orchestrator on a task. Use when the user asks to autonomously decompose and execute an engineering task with leerie.

10 5d ago A 29 tokens original MIT

judge-llm-batch

06

enricai/leerie

Skill Claude CodeCodex

Apply a 3-dimensional accuracy rubric (schema adherence, factual grounding, hallucination-freeness) to a batch of leerie LLM call captures and write verdict JSON. Used to measure whether a leerie worker (classifier, planner, implementer, etc.) is producing accurate output under its system-prompt contract.

10 5d ago A 71 tokens original MIT

llm-self-heal

07

enricai/leerie

Skill Claude CodeCodex

Autonomous self-healing loop for leerie worker prompts that produced captured failures. For each calltype with failures, runs a measured n=N baseline (unpatched), then iterates: invoke patch-generator worker (claude -p) → apply proposed patch → replay patched arm → score → check convergence…

10 5d ago A 125 tokens original MIT