A
areguig/petit-poucet
One memory for every coding agent you switch between: Claude Code, Copilot, Codex, Cursor, Antigravity
no SECURITY.md
View on GitHub ↗Link checked 2026-11-12 · HTTP 200 · opens the vendor's own site
- Star count
- 5 ★ · areguig/petit-poucet · read from the GitHub search API · (checked 2026-11-13 20:00:50)
- Source
- read directly from its GitHub page · type: open-source project
- Harvest history
- first seen 2026-10-23 · last seen 2026-11-14 · seen 26× · listed by 1 sources
- Score
- score 103 · 5 ★ · 1 sources · repo
FAQ (answers come from our records)
Who is it for?
We filed it under these tags: 记忆增强. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
areguig/petit-poucet;first catalogued 2026-10-23;last seen 2026-11-14。There is no official certification or quality rating here — we record sources and dates only.petit-poucet
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
Leaves pebbles so your coding agents find their way back.
In the French tale Le Petit Poucet, a boy drops white pebbles along the path so he can find his way home. petit-poucet does the same for AI coding agents: it keeps a small, curated memory of your rules, decisions and verified facts, and every agent you switch between starts each session with it.
Works with Claude Code, GitHub Copilot (CLI, VS Code, IntelliJ, the Copilot app), Codex, Cursor, Antigravity CLI, OpenCode and omp, on macOS, Linux and Windows.
## Why
Coding agents forget everything between sessions, and each keeps its own memory in its own format: what you taught Claude Code on Monday, Codex doesn't know on Tuesday. petit-poucet gives them one memory that:
follows you across agents: one command wires the same memory into every agent it finds;
is there from the first message: a session-start hook hands the agent your rules, so nothing depends on the model deciding to look;
is yours to read: plain Markdown in a folder you choose, browsable in Obsidian, with every change in git;
is curated: agents save a decision or a verified fact on purpose, with its source, and rules you stated change only with your confirmation;
stays clean: checks and a cleanup review catch duplicates, stale facts and unused notes.
How it compares with the agents' own memory: comparison.
## Install
sh
# macOS, Linux
curl -fsSL https://raw.githubusercontent.com/areguig/petit-poucet/main/install.sh | sh
petit-poucet setup
powershell
# Windows, in PowerShell
irm https://raw.githubusercontent.com/areguig/petit-poucet/main/install.ps1 | iex
petit-poucet setup
Then restart your agent. Already keeping memory elsewhere? Ask your agent to run the migrate-memory skill. Details, per-agent notes and settings: install and set up.
## What your agent sees
At the start of every session, the agent gets your rules and an Index of what applies here: your preferences and the current repo's notes, one line each. It opens only the notes a task needs.
🪨 petit-poucet · 20 notes loaded (preferences + chargepath-api)
markdown
## Preferences (all repos)
[[Preferences/commit-rules]] — commit locally per step, 1-2 line message, never push
[[Preferences/never-read-secrets]] — use secret values only by piping them into commands, never print them
In the French tale Le Petit Poucet, a boy drops white pebbles along the path so he can find his way home. petit-poucet does the same for AI coding agents: it keeps a small, curated memory of your rules, decisions and verified facts, and every agent you switch between starts each session with it.
Works with Claude Code, GitHub Copilot (CLI, VS Code, IntelliJ, the Copilot app), Codex, Cursor, Antigravity CLI, OpenCode and omp, on macOS, Linux and Windows.
## Why
Coding agents forget everything between sessions, and each keeps its own memory in its own format: what you taught Claude Code on Monday, Codex doesn't know on Tuesday. petit-poucet gives them one memory that:
follows you across agents: one command wires the same memory into every agent it finds;
is there from the first message: a session-start hook hands the agent your rules, so nothing depends on the model deciding to look;
is yours to read: plain Markdown in a folder you choose, browsable in Obsidian, with every change in git;
is curated: agents save a decision or a verified fact on purpose, with its source, and rules you stated change only with your confirmation;
stays clean: checks and a cleanup review catch duplicates, stale facts and unused notes.
How it compares with the agents' own memory: comparison.
## Install
sh
# macOS, Linux
curl -fsSL https://raw.githubusercontent.com/areguig/petit-poucet/main/install.sh | sh
petit-poucet setup
powershell
# Windows, in PowerShell
irm https://raw.githubusercontent.com/areguig/petit-poucet/main/install.ps1 | iex
petit-poucet setup
Then restart your agent. Already keeping memory elsewhere? Ask your agent to run the migrate-memory skill. Details, per-agent notes and settings: install and set up.
## What your agent sees
At the start of every session, the agent gets your rules and an Index of what applies here: your preferences and the current repo's notes, one line each. It opens only the notes a task needs.
🪨 petit-poucet · 20 notes loaded (preferences + chargepath-api)
markdown
## Preferences (all repos)
[[Preferences/commit-rules]] — commit locally per step, 1-2 line message, never push
[[Preferences/never-read-secrets]] — use secret values only by piping them into commands, never print them
Listed badge (optional)
Copy this line into the README if you want the listing mark. Optional, and it does not affect being listed.
[](https://anclum.com/en/s/areguig-petit-poucet/)
- Source
- github · https://github.com/areguig/petit-poucet
- Browse by field
- Memory & context
- Traction
- 5 ★
- added
- today
- Security policy
- no
SECURITY.mdat the repository root (which does not mean it is unsafe)
More like this
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.