remember

A Harmonia memory command for saving one lesson in either project memory or shared memory.

In plain words
What is it for?
Use it to record a single project-specific or cross-project learning with a title and topic tags.
Why use it?
It keeps useful knowledge available for later work while requiring you to choose where the lesson belongs and preventing client-specific information from entering shared memory.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/foliveira/harmonia/remember
Any agent
npx skills add foliveira/harmonia --skill remember
Clone the repo
git clone --depth 1 https://github.com/foliveira/harmonia

Made for: Claude Code, Codex.

Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 265 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00029 $0.00265
Opus 5 $0.00015 $0.00133
Sonnet 5 $0.00006 $0.00053
Haiku 4.5 $0.00003 $0.00026

Measured 2d ago against content hash c1a9b3fda224, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

remember 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.

skills/remember/SKILL.md · 21 lines

What it actually says

Capture ONE learning through the memory script with an explicit tier. Never a bare or defaulted global write.

  1. Elicit the tier from the developer - do not guess it:

    • project - specific to this repo (the usual choice);
    • global - a cross-project pattern, carrying NO client content. Ask which tier, and pass --client for a client-work learning.
  2. Capture it, body on stdin:

echo "<the learning body>" | bash ${CLAUDE_PLUGIN_ROOT}/bin/memory/capture.sh --tier <project|global> [--client] --title "<short title>" --tags "<lang,topic>"

Client content never reaches the global tier (R21): capture.sh refuses a --client write to --tier global with a non-zero exit, and refuses a global entry with no recognized language tag (it would be unreachable by recall). That deterministic guard is the backstop behind the tier you elicit - not a licence to default the tier. Surface the script's output.

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. 2d ago First seen · 21 lines · 29 tokens per session scan A c1a9b3fda224

Subscribe to this mod's changes

remember is a skill published in the GitHub repository foliveira/harmonia (2 stars, last pushed 17d ago), licensed MIT. It adds 29 tokens to every session and 265 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.