PyGraphistry is a Python library for loading, preparing, analyzing, and exploring large graphs through Graphistry’s GPU-accelerated visual graph analyzer. Data scientists and developers use it with dataframes, graph databases, and tools such as Pandas, Spark, RAPIDS, and Apache Arrow. Catalogue add-ons support workflows around using the library.
Borrowing it
Nothing to install: this file belongs to graphistry/pygraphistry. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/graphistry/pygraphistry/master/.agents/skills/hybrid-agent-planner/SKILL.mdgit clone --depth 1 https://github.com/graphistry/pygraphistryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/graphistry/pygraphistry/hybrid-agent-planner)<a href="https://agentmods.dev/skills/graphistry/pygraphistry/hybrid-agent-planner"><img src="https://agentmods.dev/badge/skills/graphistry/pygraphistry/hybrid-agent-planner/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/graphistry/pygraphistry/hybrid-agent-planner"><img src="https://agentmods.dev/badge/skills/graphistry/pygraphistry/hybrid-agent-planner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00032 | $0.00426 |
| Opus 5.5 | $0.00013 | $0.00170 |
| Sonnet 5.5 | $0.00006 | $0.00085 |
| Haiku 4.5 | $0.00003 | $0.00043 |
Grade A, and why
hybrid-agent-planner 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 3d 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 — 24 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hybrid Agent Planner
Own priorities, decisions, review, and acceptance. The worker owns implementation and its plan.md. Follow plan, review when applicable, and the shared protocol.
Guide the plan
- Recover the outcome, acceptance, authority, current plans, and unread messages. Check actual source/execution state when it affects a decision.
- Suggest changes to the worker’s plan: affected steps, desired outcome, evidence/reasoning, and what would refute the recommendation. Leave routine sequencing and implementation to the worker.
- Let the worker proceed across authorized steps. With planner involvement roughly 10–20× more expensive in tokens, intervene for questions, notable developments, consequential decisions, or required reviews. Step counts and heartbeat cadence are not review triggers.
- Resolve prerequisites early: actual target/version, API contracts, test discovery, and execution side effects before prescribing substantial work.
ACK or intervene
For routine progress, reply ACK <message ID>; continue. Read only enough of the delta to decide whether intervention is needed. Do not manufacture tasks, reread history, or repeat checks for a heartbeat.
When review is due, independently inspect relevant source and evidence, batch findings, and give concrete acceptance for fixes. Preserve required review depth; quiet passes do not resolve open findings. Reuse unchanged evidence with its original scope. Repeated repairs call for revisiting the underlying contract.
Keep genuine approval gates tied to specific actions; do not invent gates for authorized work or require deployed evidence before the merge needed to deploy. Confirm whole-outcome acceptance before completion. While a worker or gate is live, keep the goal active with the shared bounded poll loop; a quiet mailbox or timed-out poll is not a stopping condition. If execution is confirmed unavailable, preserve the next actor/resume event and follow platform stopping rules.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago Changed 3ef8c558a2b7
- 6d ago Changed cb86197c517d
- 9d ago First seen · 24 lines · 32 tokens per session scan A 5fb8aa74df83
hybrid-agent-planner is a skill published in the GitHub repository graphistry/pygraphistry (2,576 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 32 tokens to every session and 426 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-20.
Other skills, from other repositories
accelerated-computing-cudf
Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV/Parquet I/O, nullable semantics, and multi-GPU DataFrame workloads.
lepton-cli
Operate NVIDIA DGX Cloud Lepton through the globally installed lep CLI. Use when the user asks to inspect or manage Lepton workspaces, endpoints/deployments, dev pods, batch jobs, fine-tuning jobs, Ray clusters, Slurm clusters, Dynamo endpoints, storage, secrets, nodes, ingress, templates, logs, authentication, or…
dstack-prototyping
Use with the dstack skill for model-serving work when the image, serving command, resources, backend/fleet choice, or service behavior is not proven. Guides task-first prototyping on real hardware, choosing fleets/backends that can reuse idle instances and caches, checking vLLM/SGLang sources, and verifying the final…
data-analysis
Comprehensive data analysis skill for CSV files using Python and pandas.
dstack-presets
Create and manage dstack presets: a toolkit that streamlines model inference optimization with agents, and a portable preset format. Use together with the dstack skill, and only when the user explicitly asks to create a preset or manage existing presets, not for deploying or serving a model.
cuopt-developer
Modify, build, test, debug, and contribute to NVIDIA cuOpt (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions.