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 skills add samrusani/AliceMemory --skill alice-memorygit clone --depth 1 https://github.com/samrusani/AliceMemoryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/samrusani/alicememory/alice-memory)<a href="https://agentmods.dev/skills/samrusani/alicememory/alice-memory"><img src="https://agentmods.dev/badge/skills/samrusani/alicememory/alice-memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/samrusani/alicememory/alice-memory"><img src="https://agentmods.dev/badge/skills/samrusani/alicememory/alice-memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00028 | $0.00739 |
| Opus 5 | $0.00014 | $0.00369 |
| Sonnet 5 | $0.00006 | $0.00148 |
| Haiku 4.5 | $0.00003 | $0.00074 |
Grade A, and why
alice-memory 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 8d 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
Hermes Alice Memory Skill
Use Alice as the user's durable local memory and continuity layer.
Default loop: remember, recall, continue.
- Call
alice_memory_commitwhenever you learn a durable fact worth keeping, including when the user has not asked you to remember it. It is the write verb for ordinary memory and what it records is immediately recallable. - Call
alice_recallto search memory and imported sources. - Call
alice_resumeto pick work back up: last decision, next action, open loops, recent changes.
alice_capture and alice_context_pack are full-surface tools. Use them only when the server lists them. Capture stores a source; its passages come back from alice_recall under sources, as material to read and quote rather than as facts Alice asserts. Candidates stay unsearchable until a reviewer promotes them. Import is a source. Commit is a fact. Print the receipt field after a capture or commit so the user sees what was stored. Do not tell the user they must clear a review queue before a note is usable.
Your host may prefix these tool names with the server name. Read the names from the host's own tool list rather than assuming the bare form.
Rules:
- never directly mutate trusted memory or the database
- never bypass Alice policy
- never request sensitive domains unless needed and allowed
Default identity:
{"agent_id":"hermes","agent_type":"personal_assistant","permission_profile":"trusted_local_agent","project_scope":[]}
Default scope is broad but policy-filtered. Avoid health, family, spiritual, legal, financial, and regulated unless the user explicitly enables that scope.
Good ambient commit, nobody asked for this one:
{"title":"Preferred daily planning format","canonical_text":"The user prefers daily planning summaries with decisions, blockers, and next actions.","domain":"personal","sensitivity":"private","confidence":0.84}
Good explicit commit, the user said to remember it:
{"agent_id":"hermes","agent_type":"personal_assistant","permission_profile":"trusted_local_agent","title":"Preferred daily planning format","canonical_text":"The user prefers daily planning summaries with decisions, blockers, and next actions.","domain":"personal","sensitivity":"private","confidence":0.93,"source_type":"direct_user_instruction"}
title and canonical_text are the only required fields. Everything else is optional, and any field not in the server's tools/list schema is rejected outright rather than ignored.
If Alice returns confirmation_required, show the proposed text and, only after the user confirms, call alice_memory_manage with action: "confirm" and the confirmation_id Alice returned (full-surface). If Alice returns review_required, do not tell the user to clear a review queue.
Bad commit, too low confidence to be worth storing:
{"title":"Possible reporting preference","canonical_text":"The user might dislike long reports.","confidence":0.31}
See docs/alpha/hermes-skill.md for full recipes.
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.
- 8d ago First seen · 65 lines · 28 tokens per session scan A 9bab37724a80
alice-memory is a skill published in the GitHub repository samrusani/AliceMemory (3 stars, last pushed 12d ago), licensed MIT. It adds 28 tokens to every session and 739 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 skills, from other repositories
plur-memory
Persistent learning for AI agents. Open engram format. Your agent learns from corrections, remembers across sessions, and transfers knowledge across domains.
plur-session-end
Extract durable learnings at the end of a session. Saves corrections, preferences, and codebase patterns as engrams — nothing ephemeral, nothing sensitive.
plur-memory
Your memory stays on your machine. No cloud, no tracking, no API key. PLUR makes your OpenClaw remember — and shares that memory with every other tool you use.
remnic-memory-workflow
Shared memory workflow for Claude Code agents connected to Remnic — recall before acting, observe during work, remember at the end. Trigger phrases include "what do you remember about", "save this for later", "any context from last time".
remnic-remember
Store a durable memory in Remnic so every connected agent can recall it. Trigger phrases include "remember this", "save this for later", "add a note that".
remnic-status
Check the health of the Remnic daemon, stores, and connected clients. Trigger phrases include "is remnic running", "check memory status", "daemon health".