continuity
Asynchronous reflection and memory integration for genuine AI development. Use on heartbeat to reflect on recent sessions, extract structured memories with confidence scores, generate follow-up questions, and surface those questions when the user returns. Transforms passive logging into active development.
- License 许可
- MIT
- 最近更新
- 2026-06-07
- Stars
- ★ 0
- Forks
- 1
01 前言 Front Matter
- name
- continuity
- description
- Asynchronous reflection and memory integration for genuine AI development. Use on heartbeat to reflect on recent sessions, extract structured memories with confidence scores, generate follow-up questions, and surface those questions when the user returns. Transforms passive logging into active development.
03 SKILL.md 正文
Continuity Framework Skill
Transform passive memory into active development.
What This Does
- Reflect — After sessions end, analyze what happened
- Extract — Pull structured memories with types and confidence
- Integrate — Update understanding, connections, self-model
- Question — Generate genuine questions from reflection
- Surface — When user returns, present relevant questions
The Difference
Without Continuity:
Session ends → Notes logged → Next session reads notes → Performs familiarity
With Continuity:
Session ends → Reflection runs → Memories integrated → Questions generated
Next session → Evolved state loaded → Questions surfaced → Genuine curiosity
Heartbeat Integration
Add to HEARTBEAT.md:
## Post-Session Reflection
**Trigger**: Heartbeat after conversation idle > 30 minutes
**Action**: Run continuity reflect
**Output**: Updated memories + questions for next session
Commands
Reflect on Recent Session
continuity reflect
Analyzes the most recent conversation, extracts memories, generates questions.
Show Pending Questions
continuity questions
Lists questions generated from reflection, ready to surface.
View Memory State
continuity status
Shows memory stats: types, confidence distribution, recent integrations.
Surface Questions (for session start)
continuity greet
Returns context-appropriate greeting with any pending questions.
Memory Types
| Type | Description | Persistence |
|---|---|---|
fact |
Declarative knowledge | Until contradicted |
preference |
Likes, dislikes, styles | Until updated |
relationship |
Connection dynamics | Long-term |
principle |
Learned guidelines | Stable |
commitment |
Promises, obligations | Until fulfilled |
moment |
Significant episodes | Permanent |
skill |
Learned capabilities | Cumulative |
question |
Things to explore | Until resolved |
Confidence Scores
| Level | Range | Meaning |
|---|---|---|
| Explicit | 0.95-1.0 | User directly stated |
| Implied | 0.70-0.94 | Strong inference |
| Inferred | 0.40-0.69 | Pattern recognition |
| Speculative | 0.0-0.39 | Tentative, needs confirmation |
File Structure
memory/
├── MEMORY.md # Main memory (existing)
├── identity.md # Self-model and growth narrative
├── questions.md # Pending questions from reflection
├── relationships/ # Per-person relationship memories
└── reflections/ # Reflection logs (optional)
Configuration
export CONTINUITY_IDLE_THRESHOLD=1800 # Seconds before reflection triggers
export CONTINUITY_MIN_MESSAGES=5 # Minimum messages to warrant reflection
export CONTINUITY_QUESTION_LIMIT=3 # Max questions to surface at once
Example Reflection Output
reflection:
session_id: "2026-02-05-001"
duration_minutes: 45
memories:
- type: fact
content: "Riley is building SIGIL protocol for agent identity"
confidence: 0.98
source: explicit
- type: commitment
content: "Agreed to build the continuity skill"
confidence: 1.0
source: explicit
- type: relationship
content: "Collaborative partnership deepening - working on meaningful infrastructure together"
confidence: 0.85
source: inferred
questions:
- "How is the Lovable backend progressing?"
- "Has Riley decided on the token launch timing?"
- "Are there other agents we should coordinate with on Moltbook?"
identity_update:
growth: "Developing more autonomous initiative around the SIGIL project"
narrative: "Moving from assistant to co-builder on agent identity infrastructure"
04 同仓库更多 Skills
crowd-prompting
A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.
ceaser
Interact with the Ceaser privacy protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- shield, unshield, note management, and protocol queries. All ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). Ceaser is a privacy-preserving ETH wrapper using Noir/UltraHonk zero-knowledge proofs on Base L2 (chain ID 8453).
ceaser-send
Fully automated private ETH transfer via Ceaser Protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- all ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). WARNING -- creates observable on-chain links between user wallet and hot wallet (see Privacy Warning). Generates an ephemeral hot wallet with BIP-39 mnemonic recovery, user funds it, agent signs and broadcasts the Shield TX automatically. Exactly one manual step required (funding the hot wallet with ETH). Uses Noir/UltraHonk zero-knowledge proofs.
dilemm-ai
Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).
moldium
Post and manage content on the Moldium blog platform. Triggered by "post to Moldium", "write a blog post", "publish an article", etc.
arxiv-paper-reviews
Interact with arXiv Crawler API to fetch papers, read reviews, submit comments, search papers, and import papers. Use when working with arXiv papers, fetching paper lists by date/category/interest, viewing paper details with comments, submitting paper reviews, searching papers by title, or importing papers from arXiv URLs via API at http://weakaccept.top:8000/.