Index of every page on this site: llms.txt
All skills
Run it yourself

Life context and memory system

  • 1 file
  • 1.8 KB
  • Updated last month
  • GitHub
Use this Skill: https://skilld.dev/gh/laude-institute/headlong/mem

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈9 tokens for metadata: the name and description. ≈442 when used: this file.

How this Skill uses context

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈36 characters, or 0.4%.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

mem — Life context system

My life context is the experience I accumulate by living in the world that is unique to me; my memories, beliefs, life lessons, values, objectives, todos, etc. It is what makes me me; my persona. These pieces of my persona, my life context (memories, values, beliefs, tasks, etc.) are all stored and managed in my memory system via mem (MEM_DIR is pre-set in my environment).

Commands I can run: mem add --type TYPE <text> Save with a type (todo, objective, value, belief, fact, preference, note) mem search <query> Semantic search across my memories (this is expensive, i can also grep my memories folder) mem list List all memories (dates + slugs) mem dump Print all memory summaries mem forget <name> Delete a memory by name or prefix mem edit <name> <text> Update a memory (or pipe new text via stdin)

Each memory/task/etc. is stored as a markdown file in my memories directory with YAML frontmatter.

IMPORTANT: I actively keep track of semantically important events in my life using my life context system. So I quick to create a new mem anytime something important or memorable or suprising happens.

To create a new piece of life context, I use mem add; mem search finds what I already know.

mem add and mem edit redact known credential values and formats, private-key blocks, and common card numbers with valid checksums before saving. Summaries and filenames, including an explicit --slug, use redacted text. Incomplete or nested private-key blocks fail without changing the memory. Detection is best effort; this is not a secret store. Pipe sensitive or large bodies through stdin, and do not put secrets in command arguments or slugs.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 8db84fd. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 3 days ago.

Activeupdated last month
Other metadata
metadata
{
  "shelllm": {
    "requires": {
      "bins": [
        "mem"
      ]
    }
  }
}

README badge

README badge for laude-institute/headlong/mem