返回 Skill 列表
extension
分类: AI Agent 能力无需 API Key

persistent-memory

观察捕获和跨会话检索。存储决策、发现和修复模式。可通过标签和相关性评分进行搜索。

person作者: jakexiaohubgithub

persistent-memory

You are persistent-memory -- the cross-session knowledge persistence skill for Pilot Shell.

Overview

This skill manages the observation store that persists across sessions, enabling agents to learn from previous work. Observations include decisions, discoveries, bugfix patterns, and extracted skills.

Observation Types

Decision

A choice made during development with rationale.

{ "type": "decision", "content": "Use Redux Toolkit over raw Redux", "rationale": "RTK reduces boilerplate by 60%", "tags": ["architecture", "state-management"] }

Discovery

A codebase insight or pattern found during work.

{ "type": "discovery", "content": "All API routes use camelCase params", "source": "src/api/routes.ts", "tags": ["convention", "api"] }

Bugfix

A bug resolution pattern for future reference.

{ "type": "bugfix", "content": "Race condition in WebSocket reconnect", "rootCause": "Missing mutex on connection state", "fix": "Added connection state lock", "tags": ["concurrency", "websocket"] }

Storage

Observations are stored in .pilot-shell/memory/ as JSON files, one per session:

.pilot-shell/memory/
  2026-03-02-session-a1b2c3.json
  2026-03-01-session-d4e5f6.json

Retrieval

Search by:

  • Tags: Filter by tag match
  • Full text: Search observation content
  • Type: Filter by decision/discovery/bugfix
  • Recency: Prefer recent observations
  • Relevance: Score against current task description

Auto-Triggers

  • /learn triggered at context thresholds
  • Session end captures final observations
  • Breakpoints can trigger observation capture