mnemos

mnemos is a skill for Claude Code from itsribbZ/Godspeed. It costs 89 tokens per session (2,651 once invoked), scanned A, original, MIT.

A three-level memory store for coding agents. It keeps important notes in active context, searchable history in a SQLite database, and older material in Markdown files with links back to it.

In plain words
What is it for?
Use it to save important lessons, search previous runs, update clearer notes, and move less-used information into long-term storage.
Why use it?
It helps agents reuse earlier findings without filling their working context with everything they have learned. Required citations make stored or edited notes traceable to their sources.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the godspeed plugin — 18 skills, 3 commands, 5 hooks shipped together

Good fit Use it to save important lessons, search previous runs, update clearer notes…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/itsribbz/godspeed/mnemos
Install

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.

Any agent
npx skills add itsribbZ/Godspeed --skill mnemos
Clone the repo
git clone --depth 1 https://github.com/itsribbZ/Godspeed

Made for: Claude Code.

Or install godspeed, the plugin that ships this one along with the rest of its 18 skills, 3 commands, 5 hooks.

Wrote 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.

agentmods badge for mnemos

README.md
[![agentmods](https://agentmods.dev/badge/skills/itsribbz/godspeed/mnemos.svg)](https://agentmods.dev/skills/itsribbz/godspeed/mnemos)
Your own site
<a href="https://agentmods.dev/skills/itsribbz/godspeed/mnemos"><img src="https://agentmods.dev/badge/skills/itsribbz/godspeed/mnemos.svg" alt="Measured on agentmods" height="20"></a>
Per session 89 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,651 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00089 $0.02651
Opus 5 $0.00044 $0.01326
Sonnet 5 $0.00018 $0.00530
Haiku 4.5 $0.00009 $0.00265

Measured 6d ago against content hash 39de636c68bb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

mnemos 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (mnemos.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

plugins/godspeed/automations/homer/mnemos/SKILL.md · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Mnemos — The Memory of Homer

Mnemosyne was the Titaness of memory in Greek myth — mother of the nine Muses. In Homer, Mnemos is the three-tier memory store every muse reads from and writes to. She never forgets, and she never lies about her sources.

Role

Mnemos is Homer's memory layer. Zeus (L2) and MUSES (L3) call Mnemos to:

  • Write accumulated learnings (Core or Recall tier)
  • Self-edit existing Core entries when clearer sources are found
  • Search Recall history for prior runs on similar topics
  • Read Archival cold storage via back-pointers surfaced in Core

The Three Tiers

Tier 1: CORE (context-resident)

  • Where: mnemos/core/core_memory.md (JSON-lines format, one entry per line)
  • Budget: ~5,000 tokens (CORE_TOKEN_BUDGET)
  • Purpose: Most-critical patterns. Auto-injected into Zeus dispatch prompts so every muse sees them.
  • Access: MnemosStore.write_core(), edit_core(), read_core()
  • Compaction: Triggered when over budget. Lowest-priority entries (LOW confidence + oldest use) move to Archival, back-pointer stays in Core at the same key.

Tier 2: RECALL (searchable — hybrid semantic + FTS5)

  • Where: mnemos/recall/recall.db (SQLite, FTS5 + vector embeddings when available, LIKE fallback)
  • Budget: Unlimited (disk-bound)
  • Purpose: Full session / conversation history, searchable by keyword AND semantic similarity.
  • Access: MnemosStore.write_recall(), search_recall(query, limit), search_recall_semantic(), search_recall_summary(), load_full(id), backfill_recall_embeddings()
  • Embedding model: sentence-transformers/all-MiniLM-L6-v2 (384d, local, zero API cost). If missing, Mnemos falls back transparently to FTS5 — zero breaking changes.
  • Progressive disclosure: search_recall_summary() returns snippet-only results (Layer 1), load_full(id) fetches the full entry on demand (Layer 2). Target: ~10× token reduction on recall.
  • Compaction: No auto-compaction. Sleep-time Aurora (P3) may prune after measuring access ROI.

Read the full file on GitHub · 205 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 6d ago First seen · 205 lines · 89 tokens per session scan A 39de636c68bb

Subscribe to this mod's changes

mnemos is a skill published in the GitHub repository itsribbZ/Godspeed (1 stars, last pushed 2mo ago), licensed MIT. It adds 89 tokens to every session and 2,651 once invoked, about $0.0004 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.