Agent Séance: I made dead AI-agent runs confess how they died (with local Gemma)

说明与用途

Agent Séance 使用本地 Gemma 模型解释 AI 代理失败原因，生成引用的忏悔并输出 HTML 摘要，保持数据隐私。

问题

Failed AI‑agent runs produce logs that show what happened but not why, making debugging hard and exposing sensitive data to external services.

结果

Agent Séance produces a readable, privacy‑preserving explanation for each failed agent run, enabling developers to learn from failures without exposing data.

工作流

• 1. Point seance.py at a failed trace JSON file.
• 2. Coroner rules on cause of death (e.g., infinite loop, unread error, phantom tool, context drowning, max steps).
• 3. Local Gemma model generates confession sentences, each citing a valid [step N].
• 4. Regex pass removes uncited or invalid citations, striking them from the record.
• 5. Generate an HTML page listing each dead run with its cause and confession.

证据

Agent Séance makes the run tell you why, in the voice of the agent that died.

来源

You point it at a failed trace. A deterministic coroner rules on the cause of death first: infinite loop, unread error, phantom tool, context drowning or max steps.

来源

Finally it lays every dead run on a digital ofrenda, an HTML page with marigold-orange candles where each run gets its cause of death and its confession.

来源