Memex Retro

What it does

When your agent uses it

• Tasks that involve Note-taking

Example prompts

• “/memex-retro”

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Ask yourself: what did I learn from this task that would be useful in the future?
2. 2If nothing worth remembering, skip — not every task produces insights
3. 3For each insight, draft an atomic card (one insight per card)
4. 4Fact Hygiene Check — before writing, scan the draft for implicit context that a stranger (or future AI) couldn't decode. Ask three questions
5. 5Before writing, memex search for related existing cards
6. 6Dedup check: If an existing card already covers this insight, memex read it, then update it by appending new information (use memex write…
7. 7If it's genuinely new, write a new card with [[links]] to related cards in the prose

What it can do on your machine

Read from SKILL.md and the folder at commit 453c0e3. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • node

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Memex Retro loads about 1.6k tokens when it runs. Until then it costs about 18 tokens; SKILL.md has 541 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~18

When it runs · the whole SKILL.md, loaded when a task matches

~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from iamtouchskyer/memex at commit 453c0e3, republished under its MIT licence (© iamtouchskyer). 541 words, ~1,634 tokens.

name

memex-retro

description

Save insights from completed tasks to Zettelkasten memory.

whenToUse

After completing any task involving code changes, architectural decisions, debugging, or non-trivial problem solving. Distill what you learned into atomic…

# Define once per session — DO NOT use variable assignment ($VAR won't expand correctly in zsh)
memex() { node $HOME/.claude/plugins/cache/cc-plugins/memex/*/dist/cli.js "$@"; }

# Then use normally
memex search "some query"
memex read some-slug
memex write some-slug << 'EOF'
card content here
EOF

digraph retro {
    "Task completed" -> "Distill: what insights came from this task?";
    "Distill: what insights came from this task?" -> "Any insights worth keeping?" [shape=diamond];
    "Any insights worth keeping?" -> "Done, no card written" [label="no"];
    "Any insights worth keeping?" -> "For each insight: draft atomic card" [label="yes"];
    "For each insight: draft atomic card" -> "Fact Hygiene Check (WHO/WHAT-WHEN/RELATIONSHIP)";
    "Fact Hygiene Check (WHO/WHAT-WHEN/RELATIONSHIP)" -> "Ambiguity found?" [shape=diamond];
    "Ambiguity found?" -> "Fix draft: make implicit context explicit" [label="yes"];
    "Fix draft: make implicit context explicit" -> "Fact Hygiene Check (WHO/WHAT-WHEN/RELATIONSHIP)";
    "Ambiguity found?" -> "memex search for related existing cards" [label="no"];
    "memex search for related existing cards" -> "memex read candidates";
    "memex read candidates" -> "Existing card covers same insight?" [shape=diamond];
    "Existing card covers same insight?" -> "Update existing card (append new info)" [label="yes"];
    "Existing card covers same insight?" -> "Write new card with [[links]]" [label="no"];
    "Update existing card (append new info)" -> "memex write <existing-card>";
    "Write new card with [[links]]" -> "memex write <new-card>";
    "memex write <existing-card>" -> "More insights?" [shape=diamond];
    "memex write <new-card>" -> "More insights?" [shape=diamond];
    "More insights?" -> "For each insight: draft atomic card" [label="yes"];
    "More insights?" -> "Done" [label="no"];
}

---
title: <descriptive title>
created: <today's date YYYY-MM-DD>
source: <auto-filled by client>
category: <optional category>
---

<One atomic insight, written in your own words.>

<Natural language sentences with [[links]] to related cards, explaining WHY they're related.>

© iamtouchskyer, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/memex-retro of iamtouchskyer/memex.

Open the folder on GitHubat commit 453c0e3