Punk Records
One brain for all your AI agents. Persist every CLAUDE.md, AGENTS.md, napkin, and rules file
from every project and machine in a single versioned store. Sync it over any git remote, or carry it
across an air gap on a USB stick. Agents can read and write it live over MCP.
On Egghead, Dr. Vegapunk keeps one shared brain, the Punk Records. His satellites each go off and learn things, and whatever they learn flows back into that shared brain. This project does the same for your agents: each project and machine is a satellite, and
~/.punk-recordsis the brain.The name is a nod to One Piece (the linked wiki pages contain spoilers). This is a fan-named project, not affiliated with Eiichiro Oda, Shueisha or Toei Animation.
Why
Agent memories are scattered across dozens of repos and dotfiles. Many of them are deliberately
uncommitted (CLAUDE.local.md, napkins, ~/.claude/CLAUDE.md). They get lost when you wipe a
laptop, they diverge between your work and home machines, and they can't follow you into an
offline network.
How it compares
- Committing
CLAUDE.mdto each repo. Keep doing it for conventions your team shares. It does not cover the memories you keep out of git on purpose (CLAUDE.local.md, napkins,~/.claude), notes you would rather not push to a team repo, or a machine that cannot reach the repo host. Punk Records stores those too, alongside the committed ones. - Dotfiles managers (chezmoi, yadm, GNU Stow). They are built for files at fixed paths in your home directory. Most agent memories live inside project checkouts, at different paths on each machine. Punk Records finds them by itself, matches checkouts by their git remote, merges edits made on two machines, scans for secrets before storing, and serves everything to agents over MCP. The two work together: a global memory that is a symlink into your dotfiles repo is followed (see below).
What it picks up
| Scope | Files |
|---|---|
| Any project | CLAUDE.md, CLAUDE.local.md, AGENTS.md, AGENT.md, GEMINI.md, napkin.md, CONVENTIONS.md, .cursorrules, .windsurfrules, .clinerules, .goosehints (anywhere in the tree) |
.claude/*.md (napkins, commands, agents), .claude/skills/*, .cursor/rules/*, .windsurf/rules/*, .clinerules/*, .roo/rules/*, .kiro/steering/*, .github/copilot-instructions.md, .github/instructions/*.md |
|
Global ($HOME) |
~/.claude/CLAUDE.md, ~/.claude/napkin.md, ~/.claude/{commands,agents,skills}/*, ~/.codex/AGENTS.md, ~/.gemini/GEMINI.md, ~/.config/goose/.goosehints |
Run punk scan to see what would be captured. To add your own patterns, put globs in punk.json
inside the store ({"include": ["docs/ai/*.md"], "exclude": []}).
Install
Requires Python 3.9+ and git. No other dependencies.
pipx install punk-records # or: pip install punk-records
Quickstart
punk init # creates ~/.punk-records and tracks your global memories
punk track ~/code/app ~/code/api
punk save # snapshot every memory file into the store
punk context ~/code/app # everything an agent should know there, as one document
Online sync (any git host)
Create a private repo (GitHub, GitLab, Gitea, a bare repo on your NAS, anything git can reach):
punk git remote add origin git@github.com:you/my-punk-records.git
punk sync # save -> pull -> push -> restore
On another machine:
punk init git@github.com:you/my-punk-records.git
punk track ~/src/app # matched to the same records through the project's git remote
punk sync
Projects get their identity from their git remote (github.com/you/app), so the same repo checked out
at different paths on different machines shares one set of memories. Projects without a remote get
local/<dirname>; you can override it with punk track --id.
Encrypted remote (untrusted host)
If the host should not see your memories at all, let punk encrypt them with age
(install age first; still no Python dependencies):
punk encrypt init git@github.com:you/my-punk-records.git # makes ~/.config/punk/age.key if missing
punk sync
Your store stays plain text on each machine; the remote only ever holds one file,
punk-records.bundle.age, replaced on every push. On another machine, copy ~/.config/punk/age.key
(keep it private and backed up; without it the remote cannot be read) and run
punk init --encrypted <url>. A machine can also keep its own key: run
punk encrypt add-recipient <its age1... public key> on a machine that can already read the remote,
then sync there. punk encrypt status shows the key and recipients. If an encrypted store also has a
plain git remote, punk sync refuses to run rather than push anything unencrypted.
Air-gapped sync
punk export /media/usb/records.bundle # on the connected side
punk import /media/usb/records.bundle # on the air-gapped side (and back again)
A bundle is a standard git bundle file. It carries full history and can be verified on arrival, and
imports merge the same way online pulls do.
Use it with your AI tools
Let your agent set it up
Paste one of these into Claude Code, Cursor, Codex or any agent that can run shell commands:
Install https://pypi.org/project/punk-records with pipx, run
punk init, track this repo withpunk track ., save, and register its MCP server for Claude Code for all my projects (claude mcp add --scope user punk-records -- punk mcp). Then runpunk scanand tell me which memory files it found.
I'm setting up a new laptop. Use https://pypi.org/project/punk-records to bring back my agent memories:
punk init git@github.com:me/my-punk-records.git, then track every repo under ~/code and runpunk sync.
Use punk-records (https://pypi.org/project/punk-records) to keep this project's CLAUDE.md and AGENTS.md in sync across my machines automatically: run
punk install-hooks.
Claude Code
claude mcp add --scope user punk-records -- punk mcp # all projects; without --scope: this project only
punk install-hooks # restore at session start, save + sync at the end
There is no server to keep running: Claude Code starts punk mcp when a session begins and talks to
it over stdin/stdout. Run punk init once first. If punk is not found, use the full path from
which punk.
Cursor, Codex and other MCP clients
Cursor: add to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project):
{ "mcpServers": { "punk-records": { "command": "punk", "args": ["mcp"] } } }
Codex (CLI, IDE extension and ChatGPT desktop share this):
codex mcp add punk-records -- punk mcp # or [mcp_servers.punk-records] in ~/.codex/config.toml
Any other MCP client works the same way: a stdio server with "command": "punk" and
"args": ["mcp"].
Things to ask your agent
| You say | The agent uses |
|---|---|
| "Before you start, check punk records for what I already know about this repo." | get_context |
| "Search punk records across all my projects: how did I set up GitHub Actions caching?" | search_memories |
| "Which projects do I have memories for?" | list_satellites, list_memories |
| "Show me the napkin for github.com/me/api." | read_memory |
| "Remember in punk records: integration tests need Docker running." | append_memory |
| "Before we finish, save the gotchas you hit today to this project's CLAUDE.md in punk records." | append_memory |
| "Sync punk records." | sync |
append_memory writes to the store, commits, and updates the local file. Secrets are refused, and it
only writes memory files (CLAUDE.md, AGENTS.md, napkins, rules).
AI tools without MCP
ChatGPT, Claude.ai, Gemini on the web or a local model can still use your memories as plain text:
punk context . | pbcopy # macOS; on Linux: | xclip -selection clipboard
punk context ~/code/api > notes.md
Then start the conversation with something like:
Here are my notes for this project (global preferences first, then project rules). Follow them, and tell me if anything you do contradicts them.
Sync automatically with Claude Code
punk install-hooks # all projects (~/.claude/settings.json)
punk install-hooks --project # only this project (.claude/settings.local.json)
This adds two Claude Code hooks: when a session starts, punk
restores your global memories and the current project's (pulling first when a remote is set); when
it ends, punk saves them and syncs. Nothing is printed into the conversation unless something needs
your attention, such as a conflict or a blocked secret. Your other settings and hooks are left alone,
the previous file is kept as .punk-bak, and punk install-hooks --uninstall removes only punk's
entries. Only tracked projects are synced.
Save as you go
punk watch # save memory files a moment after they change (Ctrl+C to stop)
punk watch --sync-every 300 # ...and pull/push every 5 minutes when a remote is set
punk watch checks the memory files it knows every 2 seconds (--interval) and saves once they have
been quiet for 2 seconds (--debounce), so a burst of edits becomes one commit. New memory files are
found by a full rescan every 30 seconds (--rescan). No file-system events, no extra dependencies.
From the shell: punk remember "- integration tests need docker running" appends to the current
project's CLAUDE.md (use -f .claude/napkin.md to target another file).
How sync works
- The store is a git repo. Every save is a commit, so
punk git log -pshows how your agents' knowledge changed over time, and nothing is ever truly lost. - Three-way per file. Punk records the local and store hash of each file at its last sync, which
lets it tell "edited here" apart from "edited on another machine". When both sides changed, it
reports a conflict and leaves both copies alone.
punk diff <satellite> <path>shows what differs (-store,+local). You resolve it withsave --force(local wins) orrestore --force(store wins, and your local copy is kept as.punk-bak). - Concurrent appends merge. Memory files use git's
unionmerge driver, so when two machines add lines to the sameCLAUDE.md, the result contains both sets of lines. The catch: if both machines change the same line, the merged file keeps both versions, one after the other. Nothing is lost, but read the file after such a sync and delete the line you no longer want;punk git log -pshows which machine wrote which. When two machines add the same line in different places, both copies stay;punk dedupelists them andpunk dedupe --applykeeps only the first (blank lines, code blocks, indented lines and headings are never touched). - Deletes propagate on save. Locally removing a file that was synced removes it from the store.
Deletions coming from the store are only reported, unless you pass
restore --prune. - Symlinked memories are kept as links. If
AGENTS.md -> CLAUDE.md, the content is stored once and the link is recorded insatellites/<id>.json, then recreated on other machines. Only links that resolve to a memory file inside the same project are recorded. Links that point outside the project are never followed.punk scanandpunk savewarn about them instead. - Dotfiles work. For global memories, a link such as
~/.claude/CLAUDE.md -> ~/dotfiles/claude/CLAUDE.mdis followed: the content is stored, and restores write through the link into your dotfiles repo. Only non-executable Markdown/text files (or files with the same name as the link) inside$HOMEare followed, never anything under.git,.ssh,.gnupg,.awsand similar.
Safety
- Secret scanning. Before a file enters the store it is checked for API keys (Anthropic, OpenAI,
AWS, GitHub, Slack, Google, Stripe), private keys, and
password=...-style assignments. A matching file is blocked by default. You can pass--redactto store it with secrets replaced, or putpunk:allowon a line to mark a false positive. - Restricted writes.
restoreonly writes paths that match the memory patterns. It never writes into.git/, outside the project, through.., or through an existing symlink. Recreated symlinks must point from one memory path to another inside the project, so a tampered store or bundle can't plant hooks or scripts in your repos. - Keep the remote private. Memories describe your code and habits. On a host you do not trust,
use
punk encrypt(above): the host then sees one age-encrypted file. Encryption hides content but does not prove who wrote it; the restricted-write rules above still apply to anything synced in.
Commands
init [URL] create a store, or clone an existing one
track [DIR...] start persisting a project untrack ID [--purge]
list | scan what is tracked / what would be captured
save | restore store <-> local projects (--dry-run, --force)
status what save/restore would do
diff [ID] [PATH] unified diff of local memories vs the store
dedupe [ID] [PATH] duplicate lines left by merges (--apply removes them)
sync save, pull/push remote, restore
export | import air-gapped bundle files
search | show | context | remember
git ... run git inside the store
mcp MCP server on stdio
encrypt ... encrypted remote: init URL | add-recipient KEY | status
install-hooks sync automatically when Claude Code sessions start and end
watch save automatically when memory files change
Roadmap
- Built-in encryption (age) for untrusted remotes
punk watch: save automatically when memory files change- Semantic search and deduplication across projects
- Shared "team brain" satellites that you can subscribe to read-only
- Small web UI for browsing memory history
Development
python -m unittest discover -s tests -v
See CONTRIBUTING.md for the rules (no runtime dependencies, Python 3.9 syntax, tests for every change). Report security problems privately as described in SECURITY.md.
MIT licensed.
Metadata
Release files for punk-records 0.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| punk_records-0.2.0.tar.gz | 57.6 kB | With provenance | Details for punk_records-0.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| punk_records-0.2.0-py3-none-any.whl | Python 3 | none | any | Details for punk_records-0.2.0-py3-none-any.whl |
Total release size: 99.9 kB
Release files / punk_records-0.2.0.tar.gz
| Download URL | punk_records-0.2.0.tar.gz |
|---|---|
| Size | 57.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
8dedc0230516fba030a251f8441074ffcf31f9bcd9fe9c291fb7fd9cc73da207
|
|
BLAKE2b-256 checksum How to use checksums |
6aa396b6a925a41a047212f4cedb405ccf3d5e6d26ef8b47283b3c8c74a30ce8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 17, 2026.
Transparency logRelease files / punk_records-0.2.0-py3-none-any.whl
| Download URL | punk_records-0.2.0-py3-none-any.whl |
|---|---|
| Size | 42.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
989ae5f7535d17a88f8dda433a89332cafe619fe0814fcddc12d2785a0484e3a
|
|
BLAKE2b-256 checksum How to use checksums |
e74ea3c2d11055f7356437d646814fdddfe7ce8912d4274d5f52b15d3b5bcb27
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 17, 2026.
Transparency logSecurity for punk-records 0.2.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for punk_records-0.2.0.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 17, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.