Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/lawofall/agentcore/conversation-logsgit clone --depth 1 https://github.com/Lawofall/AgentCoreWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00055 | $0.01226 |
| Opus 5 | $0.00028 | $0.00613 |
| Sonnet 5 | $0.00011 | $0.00245 |
| Haiku 4.5 | $0.00006 | $0.00123 |
Grade A, and why
conversation-logs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
对话日志查询
何时打开（默认关）
满足任一才往下查：贴了 32-hex / UUID / 排查包；或明说查这次失败 / 巡检 / reviews。其余讨论（产品、架构、拦截、取舍）→ 不 sync、不 patrol、不 --recent。打开后只答「这次发生了什么」；驱动改动须码核 HEAD；与已确认设计冲突时日志让路。按问题选证据 → dev-process.mdc。
两套日志，不要混
| 产品 AI 日志 | Cursor IDE Agent 会话 | |
|---|---|---|
| 位置 | logs/dev.jsonl + Postgres |
agent-transcripts/（UUID.jsonl） |
| ID | 32 位 hex 无连字符 = trace_id；带连字符 UUID = conversation_id | 文件名 = UUID 带连字符 |
用户给 id → logs/dev.jsonl + DB，禁止在文件系统搜。agent-transcripts/ 与产品 AI 无关。日志不含消息正文（在 Postgres）。
查询入口（apps/server 下）
uv run python scripts/log_timeline.py --trace <trace_id> # 默认 decision_spine
uv run python scripts/log_timeline.py --pack <dir> --trace <tid> # 排查包制品
uv run python scripts/log_timeline.py --pack <dir> --full --trace <tid>
uv run python scripts/log_timeline.py --raw --trace <trace_id> # 全量仅需要时
uv run python scripts/log_timeline.py <conversation_id>
uv run python scripts/log_timeline.py --recent
uv run python scripts/log_stats.py --json
uv run python scripts/log_patrol.py --export-dir ../../logs/prod-export --since 2d # 巡检①
有 ID → 优先 decision_spine / 排查包（--pack）；--raw 仅深挖。包内必有 decision_spine.json + timeline.jsonl + meta.json；有 journal 时附 journal.redacted.jsonl + journal.summary.json（永不带原文）。失败回合 persist 后可能已有 logs/packs/<trace_id>/（journal-only：meta.json + journal.redacted.jsonl，无原文，不等 jsonl）。--trace 人读 / --json 的 meta.failure_pack 会标明这份自动存档在不在；过期删除记 journal.failure_pack_gc_expired。spine_events = turn_spine，勿与产品面混称。
最短路径（已打开时）
读本 rule CLI 即可；禁止整篇读分析指南；禁止派探子。缺正文再 --pack（必要时 --full）。单条 ID（即便夹讨论根因）→ 主 Agent 自跑 CLI，不套分相。
巡检 / reviews / 开案 / 多 trace 宽窗（仅用户明说）→ logs/reviews/README.md + STATUS.md：① 主自跑 log_patrol.py；② 按密度拆只读审阅桶（拆几飞几，禁垫桶）。写成「建议改」前须对 HEAD 打开指针文件码核；翻 STATUS ≠ 码核。巡检①用 log_patrol.py（--families / --snapshot-out / --baseline / --diff），别复制一次性扫描脚本。
症状进层（假卡 / 本机横幅 / 缺窗 / token 空…）→ 分析指南 · 症状进层（按需开节）。Token 两口径 / 事件词表 / 接缝深挖 → 同文按需开节。Dogfood 入槽 → evals/dogfood/README.md。本地无命中且像线上 → pnpm sync:logs（脱敏 turn_journal；深挖本机才 --full）后 --export-dir ../../logs/prod-export。勿再用 AGENTCORE_SSH / $AGENTCORE_HOME/logs。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 51 lines · 55 tokens per session scan A 24cc8bc16f71
conversation-logs is a cursor rule published in the GitHub repository Lawofall/AgentCore (78 stars, last pushed 2d ago), licensed MIT. It adds 55 tokens to every session and 1,226 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
code-patterns
Python code style and recurring patterns (config, logging, errors, paths).
project-overview
Core project architecture, patterns, and conventions for the AI Documentation Generator.
cursorrules
You are an AI agent building the module: {{MODULENAME}} This module is part of the Mnemosyne Neural OS ecosystem by XPACEGEMS LLC.
backend-python
Python and FastAPI standards for the backend.
core-project-context
Canonical project context and ownership rules.
dashboard-typescript
React and TypeScript standards for the operator dashboard.