OpenSquilla is a token-efficient AI agent with a shared execution loop for CLI, web, and chat interfaces. It routes requests among language models, while providing persistent memory, sandboxing, web search, embeddings, and tool handling for agent-based tasks. Its catalogue entries are skills that extend the agent's workflows.
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 skills add opensquilla/opensquilla --skill awesome-webpage-researchgit clone --depth 1 https://github.com/opensquilla/opensquillaWrote 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/opensquilla/opensquilla/awesome-webpage-research)<a href="https://agentmods.dev/skills/opensquilla/opensquilla/awesome-webpage-research"><img src="https://agentmods.dev/badge/skills/opensquilla/opensquilla/awesome-webpage-research/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/opensquilla/opensquilla/awesome-webpage-research"><img src="https://agentmods.dev/badge/skills/opensquilla/opensquilla/awesome-webpage-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00038 | $0.00758 |
| Opus 5 | $0.00019 | $0.00379 |
| Sonnet 5 | $0.00008 | $0.00152 |
| Haiku 4.5 | $0.00004 | $0.00076 |
Grade A, and why
awesome-webpage-research 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 10d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Awesome Webpage Research (mini)
Lightweight, single-pass topic research used only by AwesomeWebpageMetaSkill.
Produce a concise topic brief with a short citation list so the page planner has
factual anchors. This is not a replacement for the bundled deep-research
skill; do not invoke it for general literature reviews or multi-round
investigations.
Inputs
The caller supplies:
question: the topic, audience, language, style, and webpage context to research.
Protocol
Single round. Three steps. No iteration, no plan.json, no state file.
- Identify 3-5 focused sub-questions that the page planner needs answered to ground page sections. Cover what the topic is, why it matters, key facts, common misconceptions, and at most one stat or date anchor. Do not exceed 5 sub-questions.
- Run one bounded web-search round: at most one search query per
sub-question. Use
web-searchfor every language. Prefer recent, reputable sources. Stop at one usable source per sub-question; do not chase broader coverage and do not run a second round. - Compile a single brief (target ~300-500 words) containing:
- one paragraph topic summary
- one paragraph "key facts" with inline
[1]-style citation tags - a
Sourceslist of 3-5 entries formatted as[n] Title — URL - a final
Page anchors:line listing 3-5 short phrases that can become section headings or callouts on the webpage
Rules
- Do not iterate. One search round, one synthesis pass. Return as soon as the brief is written.
- Do not fetch full articles. Use search snippets; perform at most one quick fetch per sub-question if a snippet is unusable.
- Do not invent citations. Every
[n]must map to a source in theSourceslist. If a source cannot be cited, drop the claim instead of fabricating one. - Do not produce a multi-page literature review. Cap the brief at ~500 words.
- Do not search for images, audio, or video media here; media acquisition is handled by other meta-skill steps.
- Do not call
deep-research,summarize, or any meta-skill from this step. - If the search round returns nothing usable, prepend a single line
RESEARCH_THINto the brief and continue with a question-only summary without inventing citations or sources.
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.
- 10d ago First seen · 76 lines · 38 tokens per session scan A 11f42d24c671
awesome-webpage-research is a skill published in the GitHub repository opensquilla/opensquilla (6,940 stars, last pushed 4d ago), licensed Apache-2.0. It adds 38 tokens to every session and 758 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
torch-geometric
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
esm
Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel…
structure-prediction
Protein structure prediction from sequence. ESMFold-based, single GPU, no MSA needed. Predicts 3D structures with pLDDT confidence scores for drug discovery targets.
transformers
This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning…
ai-engineering-curriculum
Use when structured AI engineering curriculum — 382 skills + 99 prompts across 20 phases covering ML, deep learning, LLMs, agents, and production systems. Use when learning AI, building AI skills,.
diffdock
Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.