Context Kit — persistent personal context for AI agents [Show and Tell] #683
JDDavenport
started this conversation in
Show and tell
Replies: 2 comments
-
|
这是来自张越的自动回复邮件。您好,你的邮件已经收到,我会尽快给您回复。
|
Beta Was this translation helpful? Give feedback.
0 replies
-
|
已收到!祝好,刘晨飞
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Context Kit — 4 Markdown templates for persistent personal context [Show and Tell]
Context Kit — an open-source solution to context amnesia in AI workflows.
The problem: AI agents don't know who they're working for. They know the task but not the operator's decision-making style, priorities, or hard rules. You re-explain yourself every session.
4 Markdown templates (Personal Context Artifacts):
wiki.md— identity, projects, domain knowledge, relationshipsmental-models.md— decision priors (money/time/risk/effort)voice.md— communication style (10 examples + 10 anti-examples)protocols.md— hard rules treated as P05 Claude Code skills: open-loops watcher, CRM, morning briefing, session digest, voice checker
Load these at session start → AI starts context-full, not context-zero.
GitHub (MIT): https://github.com/JDDavenport/context-kit
One-command install:
curl -fsSL https://raw.githubusercontent.com/JDDavenport/context-kit/main/install.sh | bashEssay (how to build a personal AI OS): https://docs.agenttree.army/articles/personal-ai-os/
Beta Was this translation helpful? Give feedback.
All reactions