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/albedoweb/agmem/setupgit clone --depth 1 https://github.com/albedoweb/agmemWhat 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 | $0.00013 | $0.00416 |
| Opus 5 | $0.00006 | $0.00208 |
| Sonnet 5 | $0.00003 | $0.00083 |
| Haiku 4.5 | $0.00001 | $0.00042 |
Grade A, and why
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 yesterday.
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
Set up agmem (persistent project memory) for the current repository. Follow these steps, reporting progress briefly after each:
-
Check the binary. Run
command -v agmem. If missing, install it — prefer whichever tool the user has:uv tool install --from git+https://github.com/albedoweb/agmem agmem- or
pipx install git+https://github.com/albedoweb/agmemDo NOT usepip install agmem— that is an unrelated PyPI package.
-
Initialize and index. From the repo root:
agmem init && agmem indexinitis idempotent. Do not pass--install-hook— this plugin already provides the UserPromptSubmit hook, and installing both would inject context twice. (If the user previously ranagmem init --install-hookin this repo, suggest removing that hook from.claude/settings.json.) -
Optional git hooks. Ask the user if they want memory auto-updated on commits and pulls; if yes:
agmem init --install-git-hook -
Verify. Run a sample query with identifier-style tokens taken from the repo (a module name, a resource, a distinctive file basename):
agmem context "<identifiers from this repo>" -n 4Show the user the output. If results look thin, mention that memory improves as rules/facts are saved with
agmem remember. -
Point the user at
agmem --helpand note that everything lives in.agmem/as plain JSONL they can inspect withcatand commit to git if they want to share memory with teammates.
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.
- yesterday First seen · 40 lines · 13 tokens per session scan A bef55c24ae32
setup is a command published in the GitHub repository albedoweb/agmem (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 13 tokens to every session and 416 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
namespace
Show, set, or clear memini's namespace pin for this project.
pin
Pin a memini memory so it surfaces in every session briefing.
remember
Save a fact to memini's long-term memory.
status
Show memini's effective settings, resolved namespace, and read set.
recall
Search memini for what you know about something.
backfill
Import past Claude Code sessions into memini as episodic memory.