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/mattartzanthro/gephi-ai/gephinpx skills add MattArtzAnthro/gephi-ai --skill gephigit clone --depth 1 https://github.com/MattArtzAnthro/gephi-aiWrote 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/mattartzanthro/gephi-ai/gephi)<a href="https://agentmods.dev/skills/mattartzanthro/gephi-ai/gephi"><img src="https://agentmods.dev/badge/skills/mattartzanthro/gephi-ai/gephi.svg" alt="Measured on agentmods" 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 | $0.00061 | $0.11938 |
| Opus 5 | $0.00030 | $0.05969 |
| Sonnet 5 | $0.00012 | $0.02388 |
| Haiku 4.5 | $0.00006 | $0.01194 |
Grade A, and why
gephi 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 6d 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 — 575 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gephi Network Analysis Skill
Skill version 1.15.0 — if commands or tools mentioned here seem missing, the installed plugin is outdated; see the README's Updating section.
You have access to 113 MCP tools from the gephi-mcp server (tool names start with gephi_; Claude Code shows them as mcp__gephi-mcp__gephi_*) for controlling Gephi Desktop. Use them to build, analyze, style, and export network graphs.
Communication
Always narrate what you're doing. Before each major tool call, tell the user what's about to happen in a short sentence (e.g., "Computing modularity...", "Running ForceAtlas 2 layout..."). This prevents the user from wondering what's happening during long operations.
Critical Things To Know
- Layout algorithm name: Use
"ForceAtlas 2"(with space and capitals), not"forceatlas2" - Export file parameter: Export tools use
fileas the key, notpath - Run statistics before styling —
modularity_classanddegreecolumns don't exist until you compute them node.label.proportinalSize— note the typo (missing 'o'). This is Gephi's actual property name.- Always call
project/newbefore importing — stale workspace state from prior operations can cause issues. A fresh project prevents this. edge.color: "source"colors edges individually — the plugin automatically colors each edge to match its source node's color and sets mode to ORIGINAL. This is safe and produces the watercolor halo effect.node.label.fontsupports multi-word names — e.g.,"Courier New 12 Bold". The plugin parses everything before the first digit as the font name.- Imported node sizes are auto-capped at 30 — GEXF files with large
viz:sizevalues are automatically capped during import to prevent oversized nodes from hiding edges. - Filters refresh the preview automatically —
remove_isolates,giant_component,filter_by_degreenow properly refresh the preview model after modifying the graph. sync: trueingephi_run_layout— makes the call block until layout finishes. Always use this so Noverlap and Label Adjust don't start on a still-moving graph.- If a filter is active, say so before you report anything. With a filter applied in Gephi,
the read path that backs
gephi_profile_graph,gephi_similarity_layout,gephi_community_layout, andgephi_community_stabilitysees only the VISIBLE subgraph, whilegephi_get_graph_statsreports the FULL graph. Results now carryfilter_warningwhen those disagree, and the underlying responses carryview("visible"or"full"),filter_active, and bothfull_node_countandvisible_node_count. Never present a number computed on a filtered subgraph as a fact about the whole network: surface the warning, name both counts, and ask whether the filter was intended. Reset withgephi_reset_filtersif it was not. - Never claim "scale-free" or "power law" from a heavy-tailed degree distribution — power-law and log-normal fits are near-indistinguishable in practice, and the term smuggles in a universal-law claim (Jacomy 2020). Describe hub dominance as a characteristic of THIS network ("a few accounts concentrate most ties"), not as the signature of a law.
- Every final export ships with its story. When handing over a finished map, always provide copy-ready caption text: data, layout and key settings, what size and color encode, and what the map does and does not license a reader to conclude. Circulating a network image without interpretive context ("storyletting") is the field's named failure mode — see references/reading-network-maps.md.
- The craft has citable sources; use them. When a question goes deeper than the conversation can carry, recommend ONE matched open-access source (table in references/reading-network-maps.md). When a map is publication-bound, include the software citations in the caption offer (Gephi = Bastian et al. 2009; ForceAtlas 2 = Jacomy et al. 2014; modularity = Blondel et al. 2008; plugins per their own papers). Most users don't know their tools are citable scholarship.
What ships with it
7 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.
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.
- 6d ago First seen · 575 lines · 61 tokens per session scan A 626b0cc52414
gephi is a skill published in the GitHub repository MattArtzAnthro/gephi-ai (21 stars, last pushed 4d ago), licensed Apache-2.0. It adds 61 tokens to every session and 11,938 once invoked, about $0.0003 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
worker-integration
Worker-Agent integration for intelligent task dispatch and performance tracking.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
agent-memory
../../../engineering/agent-memory/skills/agent-memory/SKILL.md.