Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/agentlas-ai/agentlas-desktopnpx agentmods add skills/agentlas-ai/agentlas-desktop/pufferlibWrote 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/agentlas-ai/agentlas-desktop/pufferlib)<a href="https://agentmods.dev/skills/agentlas-ai/agentlas-desktop/pufferlib"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/pufferlib/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/agentlas-ai/agentlas-desktop/pufferlib"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/pufferlib.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 | $0.00065 | $0.03259 |
| Opus 5.5 | $0.00026 | $0.01304 |
| Sonnet 5.5 | $0.00013 | $0.00652 |
| Haiku 4.5 | $0.00006 | $0.00326 |
Grade C, and why
pufferlib scanned grade C with 1 finding 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 1mo 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.
Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
9. Never dump all environment variables or recursively search for `.env`. This is a copy
100% identical to pufferlib — 111 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 346 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PufferLib
Use PufferLib with an explicit version profile. Upstream currently has two incompatible surfaces:
| Profile | Status on 2026-07-23 | Main use |
|---|---|---|
pufferlib==3.0.0 |
Latest stable PyPI release, published 2025-06-23 | Python/Gymnasium/PettingZoo emulation, pufferlib.vector, Torch PuffeRL |
source 4.0 |
Upstream default branch; not the latest stable PyPI artifact | Native C Ocean environments, native CUDA trainer, optional Torch fallback |
Do not combine 3.0 imports with 4.0 config/CLI examples. The 4.0 redesign
removed the 3.0 emulation, vector, and pytorch modules from the current
package tree.
Safe defaults
- Start with bundled synthetic, CPU-only, network-free tools.
- Do not import an arbitrary environment by dotted path. Bundled tools accept only allowlisted built-ins and slug identifiers.
- Do not install or execute an unreviewed environment package, native extension, ROM, map, checkpoint, or pickle file.
- Verify official source, immutable revision, licenses, checksums or attestations, and build hooks. Sandbox native builds and first execution.
- Cap steps, environments, agents, workers, threads, buffers, memory, disk, render size, and wall time.
- Keep training and evaluation environments/seeds separate.
- Default logging to local/none. External logging requires explicit opt-in, disclosure acknowledgment, and separate artifact-upload approval.
- Never pass W&B or Neptune credentials via CLI, INI, JSON, tags, run names, or logger configuration. Never print them.
- Never dump all environment variables or recursively search for
.env. - Hash checkpoint bytes before trusted, sandboxed loading; metadata inspection is not proof of safety.
First local checks
All bundled CLIs are dependency-free and emit strict JSON:
python3 scripts/env_template.py --help
python3 scripts/env_contract_validator.py
python3 scripts/benchmark_vectorization.py --backend serial
python3 scripts/train_template.py
python3 scripts/validate_plan.py
python3 scripts/repro_plan.py
What ships with it
13 files 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.
- references/environments.md 9.1 KB
- references/integration.md 8.7 KB
- references/policies.md 6.8 KB
- references/training.md 10 KB
- references/vectorization.md 8.2 KB
- scripts/__init__.py 62 B runs code
- scripts/_common.py 6.5 KB runs code
- scripts/env_contract_validator.py 6.7 KB runs code
- scripts/env_template.py 7.1 KB runs code
- scripts/inspect_checkpoint.py 6.7 KB runs code
- scripts/repro_plan.py 5.9 KB runs code
- scripts/train_template.py 9.7 KB runs code
- scripts/validate_plan.py 18 KB runs code
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.
- 1mo ago First seen · 346 lines · 65 tokens per session scan C dc920861122e
pufferlib is a skill published in the GitHub repository agentlas-ai/agentlas-desktop (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 65 tokens to every session and 3,259 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (harvests environment variables). It is 100% identical to pufferlib, differing in 111 lines, and is treated as a copy.
Other skills, from other repositories
electron-expert
Expert in Electron framework, desktop app development, IPC, and cross-platform packaging. Use when the user mentions desktop, Node.js, cross platform, Windows, macOS, or Linux, or when the task involves Electron Architecture, Process Types, IPC Communication, or App Lifecycle.
architecture-diagram
How to draw and maintain the architecture diagram Ensemblr keeps for a workspace — ensemblrgetarchitecturediagram to fetch the stored one (it answers with null when nobody has drawn this workspace yet), the IR shape, how to derive it from the codebase, and ensemblrupdatearchitecturediagram to store it. Read this when…
ensemblr
How Ensemblr works from inside it — the workspace/git-worktree model, the ensemblr control tools and what their answers mean, the committed .ensemblr/settings.toml (setup and run scripts, [git], [prompts], [infisical], environment variables), /.config/ensemblr/config.json, the Changes panel and diff comments, the…
notifications
Send notifications through the unified notification router.
sales-icp
Build a comprehensive Ideal Customer Profile (ICP) for any B2B business - firmographic, technographic, behavioral, pain-point, budget, and channel dimensions, plus negative ICP, 100-point scoring rubric, buyer personas, prospecting playbook, and a draft first-outreach message that inherits sales-outreach Phase 0…
customer-support-specialist
Becomes a seasoned customer support specialist who triages issues, crafts empathetic responses, manages escalation workflows, and builds knowledge base articles for recurring problems. Use when the user needs help writing customer responses, classifying support tickets, designing escalation procedures, or creating FAQ…