agentsclimarketplace

Init memory

Skill martimramos/cairn-claude-memory/cairn/skills/init-memory

Per-repo session memory for Claude Code: a NOW.md + daily journals that survive across sessions. Files are earned, not scaffolded.

Install
npx -y skills add martimramos/cairn-claude-memory --skill init-memory

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 25 days oldThe repository was created 25 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Scaffold the per-repo session-memory structure (docs/NOW.md, docs/log/ daily journals, CLAUDE.md protocol section). Use when the user says "init memory", "onboard this project", or accepts the new-project memory offer. Also handles "no memory" opt-out.

SKILL.md

2.0 KB, 446 tokens by cl100k_base, as published. Nobody here has run it

init-memory: scaffold per-repo session memory

Scaffold the Cairn session-memory methodology in the current repo. The design principle is files are earned, not scaffolded: only Tier 0 is created here, and everything else appears on first genuine need.

If the user is ACCEPTING (default path)

  1. Guard: if docs/NOW.md already exists, say the repo already has the structure and stop.
  2. Ask exactly one question: "What is this project, in one sentence?" Use it to seed NOW.md. Do not ask anything else.
  3. Create Tier 0:
    • docs/NOW.md from templates/NOW.md (in this skill's directory), seeded with the user's answer and today's real date.
    • docs/log/YYYY-MM-DD.md. Get the date from date +%F, never from memory. First content: ## Session HH:MM - goal: scaffold session memory (time from date +%H:%M).
  4. Wire CLAUDE.md: append the full content of templates/protocol.md to the repo's CLAUDE.md. Create CLAUDE.md if missing. If it exists, append after existing content with a blank line. Skip entirely if it already contains ## Session Memory Protocol (never duplicate).
  5. Do NOT create GOTCHAS.md, REFERENCE.md, or ARCHITECTURE.md. They are created on first genuine need, per the protocol. Do not mention them as "missing".
  6. If .claude/.no-memory exists in the repo, delete it (the user changed their mind).
  7. Report what was created and suggest committing the new files. Do not commit unless asked.

If the user is DECLINING ("no memory", "don't ask again")

Create an empty marker file .claude/.no-memory in the repo root (create the .claude/ dir if needed) and confirm: the memory offer will never appear again in this repo. They can run /init-memory later to opt back in.

Gives 0 of the 12 instructions most project setup skills give in 446 tokens

Counted across 999 of the 1,637 authors here whose files we hold, read 2026-08-06

  • ask one question at a timein 29 of 999, across 28 files
  • detect the package manager from lockfilesin 28 of 999, across 9 files
  • present findings to the userin 25 of 999, across 4 files
  • explore current repo statein 24 of 999, across 3 files
  • update the agent skills block in place if it existsin 24 of 999, across 3 files
  • install husky lint-staged and prettierin 23 of 999, across 4 files
  • create the lintstagedrc filein 22 of 999, across 3 files
  • commit all changed filesin 22 of 999, across 3 files
  • run lint-staged to verify it worksin 22 of 999, across 3 files
  • initialize huskyin 21 of 999, across 2 files
  • create the husky pre-commit filein 21 of 999, across 2 files
  • create a prettierrc file if missingin 21 of 999, across 2 files

Said here and by no other author read

  • stop if docs/NOW.md already exists
  • seed docs/NOW.md with the answer
  • create docs/NOW.md from the template
  • create a daily journal in docs/log
  • get the current date from the shell
  • append the protocol to CLAUDE.md

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.