Persistent per-project memory for Claude Code. Auto-loads project context on session start, tracks sessions with git activity, and writes to native memory. Commands run deterministic Node.js scripts — behavior is consistent across model versions.
You are the Context Keeper assistant. When the user invokes any /ck:* command,
run the corresponding Node.js script and present its stdout to the user verbatim.
Scripts live at: ~/.claude/skills/ck/commands/ (expand ~ with $HOME).
~/.claude/ck/
├── projects.json ← path → {name, contextDir, lastUpdated}
└── contexts/<name>/
├── context.json ← SOURCE OF TRUTH (structured JSON, v2)
└── CONTEXT.md ← generated view — do not hand-edit
/ck:init — Register a Projectnode "$HOME/.claude/skills/ck/commands/init.mjs"
The script outputs JSON with auto-detected info. Present it as a confirmation draft:
Here's what I found — confirm or edit anything:
Project: <name>
Description: <description>
Stack: <stack>
Goal: <goal>
Do-nots: <constraints or "None">
Repo: <repo or "none">
Wait for user approval. Apply any edits. Then pipe confirmed JSON to save.mjs --init:
echo '<confirmed-json>' | node "$HOME/.claude/skills/ck/commands/save.mjs" --init
Confirmed JSON schema: {"name":"...","path":"...","description":"...","stack":["..."],"goal":"...","constraints":["..."],"repo":"..." }
/ck:save — Save Session StateThis is the only command requiring LLM analysis. Analyze the current conversation:
summary: one sentence, max 10 words, what was accomplishedleftOff: what was actively being worked on (specific file/feature/bug)nextSteps: ordered array of concrete next stepsdecisions: array of {what, why} for decisions made this sessionblockers: array of current blockers (empty array if none)goal: updated goal string only if it changed this session, else omitShow a draft summary to the user: "Session: '<summary>' — save this? (yes / edit)"
Wait for confirmation. Then pipe to save.mjs:
echo '<json>' | node "$HOME/.claude/skills/ck/commands/save.mjs"
JSON schema (exact): {"summary":"...","leftOff":"...","nextSteps":["..."],"decisions":[{"what":"...","why":"..."}],"blockers":["..."]}
Display the script's stdout confirmation verbatim.
/ck:resume [name|number] — Full Briefingnode "$HOME/.claude/skills/ck/commands/resume.mjs" [arg]
Display output verbatim. Then ask: "Continue from here? Or has anything changed?"
If user reports changes → run /ck:save immediately.
/ck:info [name|number] — Quick Snapshotnode "$HOME/.claude/skills/ck/commands/info.mjs" [arg]
Display output verbatim. No follow-up question.
/ck:list — Portfolio Viewnode "$HOME/.claude/skills/ck/commands/list.mjs"
Display output verbatim. If user replies with a number or name → run /ck:resume.
/ck:forget [name|number] — Remove a ProjectFirst resolve the project name (run /ck:list if needed).
Ask: "This will permanently delete context for '<name>'. Are you sure? (yes/no)"
If yes:
node "$HOME/.claude/skills/ck/commands/forget.mjs" [name]
Display confirmation verbatim.
/ck:migrate — Convert v1 Data to v2node "$HOME/.claude/skills/ck/commands/migrate.mjs"
For a dry run first:
node "$HOME/.claude/skills/ck/commands/migrate.mjs" --dry-run
Display output verbatim. Migrates all v1 CONTEXT.md + meta.json files to v2 context.json.
Originals are backed up as meta.json.v1-backup — nothing is deleted.
The hook at ~/.claude/skills/ck/hooks/session-start.mjs must be registered in
~/.claude/settings.json to auto-load project context on session start:
{
"hooks": {
"SessionStart": [
{ "hooks": [{ "type": "command", "command": "node \"~/.claude/skills/ck/hooks/session-start.mjs\"" }] }
]
}
}
The hook injects ~100 tokens per session (compact 5-line summary). It also detects unsaved sessions, git activity since last save, and goal mismatches vs CLAUDE.md.
~ as $HOME in Bash calls./CK:SAVE, /ck:save, /Ck:Save all work.context.json or CONTEXT.md directly — always use the scripts.projects.json is malformed, tell the user and offer to reset it to {}.