How do I install the Team Memory skill?

Run npx skillmds@latest add wodsmith/team-memory in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Team Memory safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Team Memory?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Team Memory free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Team Memory?

wodsmith (@wodsmith) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Team Memory by wodsmith — Agent Skill

Name: Team Memory
Author: wodsmith

Team Memory

Store, search, and manage team memory from AI coding sessions. Use /remember to save observations (conventions, gotchas, debugging tips, architecture decisions, workflows). Use /recall to search memories by semantic query. Use /feedback to rate memory quality (helpful/harmful/irrelevant). Triggers on any request to save, recall, search, or rate team knowledge.

wodsmith eaf004c 10 files · 22.3 KB Updated Sep 17, 2026

File contents

Team Memory

Manage a shared team memory system backed by a Cloudflare Worker with semantic search.

Commands

/remember — Store an observation

Save a new observation to team memory.

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "<observation text>" [--category=<category>] [--priority=<priority>]

Categories: convention, gotcha, debugging, architecture, workflow Priorities: critical, moderate, ephemeral

Defaults: category=convention, priority=moderate

Examples:

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "PlanetScale returns strings for COUNT columns — always wrap with Number()" --category=gotcha --priority=critical
TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "Use ULID for ID generation, not CUID2" --category=convention

/recall — Search memories

Search team memory by semantic query.

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "<query>" [--limit=<n>] [--category=<category>] [--priority=<priority>]

Default limit: 5. Results are ranked by relevance.

Examples:

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "database migration patterns"
TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "Stripe" --category=gotcha --limit=10

/feedback — Rate a memory

Provide feedback on a memory's usefulness.

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts <observation-id> <signal> [--note="<reason>"]

Signals: helpful, harmful, irrelevant

Examples:

TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts 01J5K3M2N7 helpful --note="Saved me from a production bug"
TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts 01J5K3M2N7 irrelevant

Environment

Production URL: https://team-memory.zacjones93.workers.dev

All commands below include the env var inline. If TEAM_MEMORY_URL is already set in your shell, the prefix is optional.

Hooks

SessionStart: Reminds the agent about the memory system and how to use /recall for task-relevant lookups
SessionEnd: Reads transcript JSONL, extracts user/assistant messages, POSTs to /sessions
Export: Syncs memories to MEMORY.md between  /  markers

---name: team-memorydescription: Store, search, and manage team memory from AI coding sessions. Use /remember to save observations (conventions, gotchas, debugging tips, architecture decisions, workflows). Use /recall to search memories by semantic query. Use /feedback to rate memory quality (helpful/harmful/irrelevant). Triggers on any request to save, recall, search, or rate team knowledge.---# Team MemoryManage a shared team memory system backed by a Cloudflare Worker with semantic search.## Commands### /remember — Store an observationSave a new observation to team memory.```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "<observation text>" [--category=<category>] [--priority=<priority>]```**Categories:** `convention`, `gotcha`, `debugging`, `architecture`, `workflow`**Priorities:** `critical`, `moderate`, `ephemeral`Defaults: category=convention, priority=moderateExamples:```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "PlanetScale returns strings for COUNT columns — always wrap with Number()" --category=gotcha --priority=criticalTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/remember.ts "Use ULID for ID generation, not CUID2" --category=convention```### /recall — Search memoriesSearch team memory by semantic query.```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "<query>" [--limit=<n>] [--category=<category>] [--priority=<priority>]```Default limit: 5. Results are ranked by relevance.Examples:```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "database migration patterns"TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/recall.ts "Stripe" --category=gotcha --limit=10```### /feedback — Rate a memoryProvide feedback on a memory's usefulness.```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts <observation-id> <signal> [--note="<reason>"]```**Signals:** `helpful`, `harmful`, `irrelevant`Examples:```bashTEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts 01J5K3M2N7 helpful --note="Saved me from a production bug"TEAM_MEMORY_URL=https://team-memory.zacjones93.workers.dev bun run .claude/skills/team-memory/scripts/feedback.ts 01J5K3M2N7 irrelevant```## EnvironmentProduction URL: `https://team-memory.zacjones93.workers.dev`All commands below include the env var inline. If `TEAM_MEMORY_URL` is already set in your shell, the prefix is optional.## Hooks- **SessionStart**: Reminds the agent about the memory system and how to use /recall for task-relevant lookups- **SessionEnd**: Reads transcript JSONL, extracts user/assistant messages, POSTs to `/sessions`- **Export**: Syncs memories to MEMORY.md between `` / `` markers

Frequently asked questions

Team Memory

File contents

Team Memory

Commands

/remember — Store an observation

/recall — Search memories

/feedback — Rate a memory

Environment

Hooks

Related skills

Frequently asked questions

How do I install the Team Memory skill?

What does the Team Memory skill do?

Is Team Memory safe to use?

Which AI agents work with Team Memory?

Is Team Memory free to use?

Who published Team Memory?