Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/goldenwo/universal-memory/um-setupgit clone --depth 1 https://github.com/goldenwo/universal-memoryWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00018 | $0.00425 |
| Opus 5 | $0.00009 | $0.00212 |
| Sonnet 5 | $0.00004 | $0.00085 |
| Haiku 4.5 | $0.00002 | $0.00042 |
Grade A, and why
um-setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run the plugin's first-run setup: verify a universal-memory server (health check + an authed write probe) and, on success, write ~/.um/endpoint and ~/.um/auth-token (mode 600) so every hook resolves it.
Steps:
- Ask the user for their UM server URL (default
http://localhost:6335for a local Docker install; a remote server looks likehttp://your-host:6337). - Ask whether the server needs an auth token. Loopback installs usually don't; anything reached through a tunnel/proxy does. If a token is needed and the user prefers not to paste it in chat, tell them to run the script themselves in a terminal — it has a hidden token prompt:
bash "${CLAUDE_PLUGIN_ROOT}/hooks/um-setup.sh" - Otherwise execute via the Bash tool (token flag only when one was given):
bash "${CLAUDE_PLUGIN_ROOT}/hooks/um-setup.sh" --endpoint <url> [--token <token>] - Relay the result. On failure the script prints one actionable message per cause and writes nothing — the important distinctions:
unreachable— wrong URL, server down, or tunnel/firewall in between.writes are DISABLED (HTTP 403)— the server needsUM_MCP_WRITE_ENABLED=true(andUM_MOUNT_MODE=rw) in its.env, then a restart.server is too old (404 on the write probe)— the server predates the/apicapture routes; it must be upgraded to v1.7.0 or newer.rejected the token (401)— token mismatch with the server'sUM_AUTH_TOKEN.
On success, hooks work immediately in the next session — no restart of Claude Code required for capture; the state-of-play banner appears at the next SessionStart.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 21 lines · 18 tokens per session scan A 8d6cf7e911ac
um-setup is a command published in the GitHub repository goldenwo/universal-memory (2 stars, last pushed 4d ago), licensed MIT. It adds 18 tokens to every session and 425 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
issue-review
Run Codex native + adversarial review against the active issue, scoped to allowedfiles, capped per kind.
wiring-check
End-of-task wiring gate — verify every change is connected end-to-end across kipi plugins, hooks, MCP tools, agents, bus files, canonical, and rules. Nothing dangling.
issue-closeout
Triage Codex findings via per-finding dispositions, record findingstriaged, close the active issue.
prd-personas
Run the Skeptic persona session against the active draft PRD.
prd-triage
Triage pending findings on the active PRD.
linear-drain
Create the queued Linear projects and issues that shell scripts captured offline.