Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/kanaries/pygwalker/agents-mdgit clone --depth 1 https://github.com/Kanaries/pygwalkerWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.02990 | $0.02990 |
| Opus 5 | $0.01495 | $0.01495 |
| Sonnet 5 | $0.00598 | $0.00598 |
| Haiku 4.5 | $0.00299 | $0.00299 |
Grade A, and why
pygwalker AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 216 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — PyGWalker contributor & agent guide
This is the fast-start map of the PyGWalker repo for both human contributors and coding agents. It explains how the project is put together, how to run it in dev mode with live frontend reload, and where all the logs go. Read this first — it is written to save you from re-deriving the architecture by grepping.
Deeper references:
docs/ARCHITECTURE.md(how it is built),docs/DEVELOPMENT.md(dev workflow + troubleshooting),docs/CONTRIBUTING.md(validation & CI).
1. What PyGWalker is (30-second model)
PyGWalker turns a pandas / polars / pyarrow dataframe into an interactive Graphic Walker UI inside notebooks, Streamlit, and plain web servers. It has two halves that ship together:
- Python package (
pygwalker/) — public API (walk,render,table,Walker,to_html), data parsing, and the transports that talk to the UI. - Frontend app (
app/, React + Vite) — the UI. It is compiled into JavaScript bundles that are checked into the wheel underpygwalker/templates/dist/and loaded by the Python side at render time.
The Python side never renders charts itself; it hands built JS + serialized data to a notebook/browser and then answers data/spec requests over a message channel.
2. Repo map
| Path | What lives here |
|---|---|
pygwalker/api/ |
Public entry points. adapter.py picks jupyter vs webserver; jupyter.py = notebook dispatch; walker.py = the reusable Walker; pygwalker.py = the core PygWalker. |
pygwalker/services/ |
Rendering + display. anywidget_widget.py (default transport), render.py + templates/*.html (iframe transport), global_var.py (runtime globals), jupyter_display.py. |
pygwalker/communications/ |
Kernel⇄frontend transports: anywidget_comm.py (default), hacker_comm.py (iframe), streamlit_comm.py, gradio_comm.py, reflex_comm.py. protocol.py is the shared message schema. |
pygwalker/data_parsers/ |
Dataframe/connector adapters (pandas, polars, pyarrow, SQL, spark…). |
pygwalker/templates/dist/ |
Build output (git-ignored). The JS bundles the Python side loads. |
pygwalker/utils/ |
Helpers: frontend_assets.py (locate/load bundles), log.py (logging), encoders. |
app/src/ |
Frontend source. index.tsx = entry; utils/communication.tsx = transports; dataSource/ = data ingest; interfaces/comm.generated.ts = generated protocol types; store/ = MobX state. |
scripts/ |
dev.py (dev orchestrator), compile.sh (build frontend), local_ci.py (mirror CI), generate_comm_protocol_ts.py (regenerate protocol types). |
tests/ |
Python tests + *.ipynb notebooks run by nbmake. app/tests/ holds Playwright smoke tests. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 216 lines · 2,990 tokens per session scan A a084b376225d
pygwalker AGENTS.md is an instructions file published in the GitHub repository Kanaries/pygwalker (15,953 stars, last pushed 22d ago), licensed Apache-2.0. It adds 2,990 tokens to every session, about $0.0149 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
mcp-data-explore CLAUDE.md
Claude Code instructions for Mingwei2/mcp-data-explore, covering claude.md, project overview, development commands, or and mcp server architecture.
streamlit copilot-instructions.md
Copilot instructions for streamlit/streamlit, covering streamlit repo overview, tech stack, folder structure, shell & build policy and make commands.
semiotic CLAUDE.md
Claude Code instructions for nteract/semiotic, covering semiotic claude code instructions and claude code notes.
IoTSharp AGENTS.md
Instructions for IoTSharp/IoTSharp, covering agents.md, current mission, current priority order, current concept realignment and ai workbench direction.
tableau-mcp CLAUDE.md
Instructions for tableau/tableau-mcp, covering claude.md, build & development commands, building, running the server and testing.
neko-master CLAUDE.md
Instructions for foru17/neko-master: All repo guidance lives in AGENTS.md — commands, conventions, key contracts, and the project map. Read it first.