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 skills/graphistry/pygraphistry/plannpx skills add graphistry/pygraphistry --skill plangit clone --depth 1 https://github.com/graphistry/pygraphistryWhat 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 | $0.00038 | $0.01406 |
| Opus 5 | $0.00019 | $0.00703 |
| Sonnet 5 | $0.00008 | $0.00281 |
| Haiku 4.5 | $0.00004 | $0.00141 |
Grade A, and why
plan 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 yesterday.
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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
File-Based Task Plan
Default: Use Platform Planning
Unless explicitly requested ("create a plan.md", "use the plan template"), prefer your AI platform's built-in planning. This file-based approach is only justified for multi-session or handoff work.
Use When
- User explicitly requests it
- Work spans multiple sessions / context resets
- Handoff to another AI or human
- Complex multi-PR coordination
Setup
- Copy template →
plans/[task_name]/plan.md - Replace
[placeholders] - Fill Context sections
- Mark Step 1
🔄
⚠️ gitignore: plans/ is local only — never commit unless user explicitly requests (git add -f).
Writing Style
Terse but recoverable — every token earns its place, but a cold-start agent with zero context must still understand and resume. Abbreviate, omit filler. Test: would a stranger understand this with no other context? If yes, cut further. If no, add the minimum needed.
Emoji: only use where it saves characters or adds clarity not already present in the text (e.g. status icons replace words, ⚠️ flags a genuine warning). Do not add emoji just to have them — if removing one loses nothing, remove it.
Critical Meta-Goals
This plan MUST be:
- Self-describing — all context to resume is IN THIS FILE and files it points to
- Write before run — record planned action in file before executing
- Update before continuing — record results immediately after, before next step
- Single source of truth — if not in the plan, it didn't happen
- Safe to resume — any agent can pick up by reading only this file
⚠️ External memory is unreliable. This file — and files it points to — are your ONLY memory.
Anti-Drift Protocol
Three Commandments
- RELOAD before every action — memory wiped, plan is all you have
- UPDATE after every action — unwritten = didn't happen
- TRUST ONLY THE PLAN — not memory, not assumptions
Rules
- 🚫 No assumptions — plan is truth
- 🚫 No offroading — if not in plan, don't do it
- 🔐 No secrets — never write passwords, tokens, API keys, or credentials; use
$ENV_VAR,<redacted>, or pointer to env file (e.g.source .env.local) - Parallel subagents: before spawning, mark claimed step
🔄(🤖agent_<id>_step_<N>)and record subagent plan path. Subagents must NOT edit this file — main agent updates it from their output. - Subagent plans: each subagent has its own full plan at
plans/[task]/subagents/agent_<id>_step_<N>/plan.md, same protocol. Subordinate to this file.
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.
- yesterday First seen · 131 lines · 38 tokens per session scan A 1cb350f6f9fb
plan is a skill published in the GitHub repository graphistry/pygraphistry (2,551 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 38 tokens to every session and 1,406 once invoked, about $0.0002 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 skills, from other repositories
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.
httpstat
Diagnose website and API performance using httpstat — a curl wrapper that visualizes HTTP timing breakdowns (DNS, TCP, TLS, server processing, content transfer). Use this skill whenever the user wants to debug slow websites, analyze HTTP/HTTPS latency, profile API response times, understand curl timing output, find…
geopandas
Use when performing vector spatial data analysis in Python — reading/writing shapefiles, spatial joins, overlay operations, choropleth maps. GeoPandas: extends pandas DataFrames with geometry columns for Pythonic spatial analysis.