cucuwang

AIASE 2026 Bug Hunter — probe-based crash + mismatch detection with AST smell analysis. Balanced precision/recall for pairwise evaluation.

Produces Python code from task descriptions (max 500 S-LOC)

Minimal smoke-test skill — echoes a greeting in the AIASE 2026 output contract. Use to verify the Hermes ↔ LiteLLM ↔ skill pipeline is wired up correctly.

Calculates carbon emissions from natural-language descriptions (Scope 1+2, Taiwan)

Converts natural-language questions + SQLite DDL into verified SQL queries

Use when previewing optional llms.txt proposal files, candidate JSON-LD, or crawler-control suggestions for a website build

Use when auditing a website or build output for deterministic content-readiness regressions and evidence-bounded discovery checks

Use when proposing evidence-bounded readability and structure edits without inventing content, ranking claims, or structured data

Translate vague visual feelings or screenshots into precise code changes. The AI locates the code, translates your description into technical root causes you can pick from, then modifies only after confirmation. Use /grill-me-fix when you want to change…