goldenwo

25 mods across 2 repositories, 3 stars between them.

universal-memory

02

goldenwo/universal-memory

Plugin Claude Code

Cross-session, cross-device memory for Claude Code. Markdown-first captures, vector-backed recall. Points at any universal-memory server — local Docker, remote self-hosted, or cloud-backed. Auto-starts the server on demand when configured to.

2 3d ago A tokens not measured original MIT

um-setup

05

goldenwo/universal-memory

Command

First-run setup — point the plugin at your universal-memory server, verify it, save config.

2 3d ago A 18 tokens original MIT

SessionStart

06

goldenwo/universal-memory

Hook

Runs when a session starts on startup, clear and compact, executing session-start.sh via bash. From goldenwo/universal-memory.

2 3d ago A tokens not measured copy · 63% MIT

UserPromptSubmit

07

goldenwo/universal-memory

Hook

Runs when you submit a prompt, before the agent sees it, executing user-prompt-submit.sh via bash. From goldenwo/universal-memory.

2 3d ago A tokens not measured copy · 58% MIT

Stop

08

goldenwo/universal-memory

Hook

Runs when the agent finishes a response, executing stop.sh via bash. From goldenwo/universal-memory.

2 3d ago A tokens not measured original MIT

SessionEnd

09

goldenwo/universal-memory

Hook

Runs when a session ends, executing session-end.sh via bash. From goldenwo/universal-memory.

2 3d ago A tokens not measured copy · 53% MIT

universal-memory

10

goldenwo/universal-memory

MCP server Claude CodeCodexCursor +2

MCP server "universal-memory", hosted remotely at localhost, as configured in goldenwo/universal-memory.

2 3d ago A tokens not measured original MIT

claude-state-drift

12

goldenwo/claude-state-drift

Plugin Claude Code

A per-project state layer for long Claude Code sessions: keeps a .claude/state.json (objective / current focus / deliverables), prints a WHERE YOU ARE orientation block at session start, re-injects the goal as you work, and nudges when state goes stale — with /where-am-i, /update-state, and /re-anchor.

1 14d ago A tokens not measured original MIT

clean

13

goldenwo/claude-state-drift

Command

Keep state.json lean by archiving old done deliverables -- dry-run first, confirm, then apply. Use when state.json feels bloated or the state-bloat nudge fires.

1 14d ago A 37 tokens original MIT

handoff

14

goldenwo/claude-state-drift

Command

Compose an end-of-session handoff -- state.json update (with approval) first, then a narrative handoff the next session in this project auto-loads. Use at session end, when context runs high, or after the context-pressure nudge fires.

1 14d ago A 51 tokens original MIT

re-anchor

15

goldenwo/claude-state-drift

Command

Drift mitigation — re-read project objective from .claude/state.json and audit the current session for alignment. Use when you sense scope drift, after long stretches of work, or before declaring a task done. Reports an alignment audit; does not modify state.

1 14d ago A 52 tokens original MIT

stats

16

goldenwo/claude-state-drift

Command

Show this project's claude-state-drift telemetry — sessions, per-injection token cost, activity counts, nudge-to-update conversion, and staleness-resolution — computed locally from the opt-in hook log. Use when you want to see what the plugin is doing and what it costs, or to check your own numbers.

1 14d ago A 65 tokens original MIT

update-state

17

goldenwo/claude-state-drift

Command

Draft an update to .claude/state.json based on recent work and show the diff. Use when substantial progress has been made (deliverable shipped, new work started, scope deferred, open question resolved), or when the user says "update state", "log progress", "snapshot where we are", or "checkpoint". Never auto-writes …

1 14d ago A 78 tokens original MIT

where-am-i

18

goldenwo/claude-state-drift

Command

Print a "WHERE YOU ARE" orientation block for the current project — objective, version, current focus, in-progress deliverables, deferred items, blocked items, open questions, recent commits, and recent snapshots. Use when you need a quick project-state overview at the start of a session or after context switches.

1 14d ago A 62 tokens original MIT

PostToolUse

19

goldenwo/claude-state-drift

Hook

Runs after a tool call finishes for Bash tool calls, executing state-track-commit.sh via bash. From goldenwo/claude-state-drift.

1 14d ago A tokens not measured original MIT

UserPromptSubmit

20

goldenwo/claude-state-drift

Hook

Runs when you submit a prompt, before the agent sees it, executing focus-check.sh via bash. From goldenwo/claude-state-drift.

1 14d ago A tokens not measured copy · 53% MIT

Stop

21

goldenwo/claude-state-drift

Hook

Runs when the agent finishes a response, executing state-staleness.sh via bash. From goldenwo/claude-state-drift.

1 14d ago A tokens not measured original MIT

SessionStart

22

goldenwo/claude-state-drift

Hook

Runs when a session starts, executing session-start-orient.sh via bash. From goldenwo/claude-state-drift.

1 14d ago A tokens not measured copy · 55% MIT

handoff

23

goldenwo/claude-state-drift

Skill Claude CodeCodex

Compose an end-of-session handoff — updates state.json first (with approval), then writes a narrative handoff that the next session in this project auto-loads. Use at session end, when context runs high, when the user says 'write a handoff', 'wrap up', 'hand off', or after the context-pressure nudge fires.

1 14d ago A 73 tokens original MIT

re-anchor

24

goldenwo/claude-state-drift

Skill Claude CodeCodex

Drift-mitigation skill. Re-reads the project's master objective from .claude/state.json and audits the current session's recent work for alignment. Use after a long stretch (>30 minutes of intensive work, or many tool calls without checking back), when noticing scope drift, before declaring a task done, or when the…

1 14d ago A 108 tokens original MIT