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 zjunlp/SciAtlas --skill sciatlas-idea-generategit clone --depth 1 https://github.com/zjunlp/SciAtlasWrote 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/zjunlp/sciatlas/sciatlas-idea-generate)<a href="https://agentmods.dev/skills/zjunlp/sciatlas/sciatlas-idea-generate"><img src="https://agentmods.dev/badge/skills/zjunlp/sciatlas/sciatlas-idea-generate/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/zjunlp/sciatlas/sciatlas-idea-generate"><img src="https://agentmods.dev/badge/skills/zjunlp/sciatlas/sciatlas-idea-generate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 14 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00092 | $0.01601 |
| Opus 5 | $0.00046 | $0.00800 |
| Sonnet 5 | $0.00018 | $0.00320 |
| Haiku 4.5 | $0.00009 | $0.00160 |
Grade A, and why
sciatlas-idea-generate 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 9d 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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SciAtlas Idea Generate
Use this skill to generate research ideas from SciAtlas KG evidence. Run only the repository workflow python -m sciatlas_idea_gen.main, which retrieves anchor papers, builds a research graph, searches same-field and cross-domain inspirations, generates ideas, and runs novelty checks.
Operating Contract
- Own the end-to-end novice flow: install or locate the workflow, guide registration, configure
.envor shell variables, run retrieval/generation, inspect artifacts, and synthesize final idea seeds. - Ask the user only for human-only values: missing topic/domain, email, verification code, SciAtlas token, LLM key/base/model, S2 key, local KG credentials when explicitly using local KG, or one necessary scope clarification.
- Do not ask the user to run shell commands when tool access is available.
- Never disclose the full API token.
- Use
sciatlas_idea_genas the only idea-generation workflow. - If
sciatlas_idea_genis unavailable or required LLM/KG setup is missing, report the blocker and fix setup if possible; do not substitute another route. - Save and inspect artifacts before writing the final answer.
KG Linkage
The workflow reaches the KG through sciatlas_idea_gen.clients.SciAtlasClient:
- Hosted mode:
SCIATLAS_USE_LOCAL_KG=0; the adapter shells into this repository'srun_sciatlas.pyfor KG-backed paper retrieval, then normalizesdata.result.ranking.papersordata.result.sources.kg.paperstopapers. - Local KG mode:
SCIATLAS_USE_LOCAL_KG=1; the adapter runsreferences/search/run_search.pywith--disable-s2,--disable-llm-ranking,--kg-top-k, and--final-top-k, then normalizes the local KG response topapers. - Keep hosted mode as the default unless the user says they have Neo4j and local KG models configured.
Zero-Start Bootstrap
- Check for the repository workflow:
python -m sciatlas_idea_gen.main -h
- If missing, clone the full SciAtlas repository, change into it, then run
python -m pip install -e ./sciatlasandpython -m pip install -r requirements-workflows.txt. Do not use the GitHub#subdirectory=sciatlaspackage-only installation: it does not includesciatlas_idea_gen. - Check current environment and
.envfor hosted KG and LLM settings before asking the user. - If
SCIATLAS_API_KEYis missing for hosted mode, guide the user tohttp://sciatlas.openkg.cn/register; ask for email, verification code, and returnedsciatlas_xxxtoken only when needed. - Configure hosted KG yourself:
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.
- 9d ago First seen · 132 lines · 92 tokens per session scan A b7cfc618fd96
sciatlas-idea-generate is a skill published in the GitHub repository zjunlp/SciAtlas (146 stars, last pushed 1mo ago), licensed MIT. It adds 92 tokens to every session and 1,601 once invoked, about $0.0005 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
jurisd-research
Expert Australian/NZ legal research and AGLC4 citation using the jurisd MCP server. Use when finding cases or legislation (AustLII), looking up a provision offline, formatting or resolving citations, building a pinpoint, tracing who-cites-what, or producing an AGLC4 bibliography. Triggers on case law, legislation…
idea-spark
Generate ONE reviewer-defensible, implementable research idea with a concrete method and falsification plan from a stated research direction. Use when the user asks for a research idea, novelty analysis, bottleneck diagnosis, or paper-shape suggestion. Skip code review, debugging, and unconstrained brainstorming…
paper2reel
Build an interactive HTML viewer that aligns paper2poster output with slide/video deck frames through a sidecar contentalignment.json, without modifying the original poster, PPTX, or blog deliverables.
graph-mutation-plan
Cookbook for composing an applygraphmutations plan — stable entitykey patterns, the canonical label/edge vocabulary, evidence/invalidation/confidence discipline, and a worked example. Load this when building a non-trivial mutation plan.
potpie-infra-architecture
Use for project infra and architecture context: environments, adapters, runtime configuration, deployments, service dependencies, datastores, API contracts, ownership, incidents, and dependency blast radius.
alfworld-search-pattern-executor
Systematically searches a sequence of likely locations for a target object based on common sense. Use when you need to find a specific object and know which receptacles to check but not which one contains it. Takes a list of candidate receptacles, orchestrates navigation and inspection, and outputs when the target is…