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/arcadi4/nerdy/maximum-flownpx skills add Arcadi4/nerdy --skill maximum-flowgit clone --depth 1 https://github.com/Arcadi4/nerdyWhat 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 | $0.00059 | $0.03824 |
| Opus 5 | $0.00030 | $0.01912 |
| Sonnet 5 | $0.00012 | $0.00765 |
| Haiku 4.5 | $0.00006 | $0.00382 |
Grade A, and why
maximum-flow 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 2d 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 — 298 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maximum Flow
Overview
Use this skill to decide whether a problem is really a flow model, a cut certificate, an augmenting-path algorithm question, or a matching reduction before naming an algorithm.
Core principle: a maximum-flow answer is correct only when the model preserves capacity constraints, conservation, residual cancellation, and the cut certificate that proves optimality.
Shared CLRS Conventions
- Follow the parent
clrsskill for mathematical formatting: every formula, inequality, asymptotic bound, and symbolic expression belongs in a display LaTeX block. - Use chapter-specific theorem names here, and keep shared graph traversal mechanics in
elementary-graph-algorithms. - State the graph direction, source and sink convention, capacity domain, and whether integrality matters before using Ford-Fulkerson, Edmonds-Karp, or a matching reduction.
- Keep tables verbal. Put capacities, residual definitions, and running times in display blocks near the table instead of inside table cells.
When to Use
Use this skill when a task involves:
- modeling transport, assignment, disjoint paths, escape routes, image segmentation-style cuts, or feasibility through capacities;
- proving a flow is maximum or extracting a minimum cut from a residual network;
- choosing or analyzing Ford-Fulkerson, Edmonds-Karp, capacity scaling, or push-relabel-style production alternatives;
- reducing bipartite matching, vertex capacities, multiple sources, multiple sinks, or antiparallel directed edges to ordinary single-source single-sink flow;
- reviewing an implementation that maintains residual edges, reverse edges, bottlenecks, or integer-flow assumptions.
Do not use this skill merely because a graph is directed. If the task is shortest paths, reachability, strongly connected components, spanning trees, or topological order without capacities and conservation, route to elementary-graph-algorithms, shortest-paths, or minimun-spanning-trees as appropriate instead.
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.
- 2d ago First seen · 298 lines · 59 tokens per session scan A e2f4713172a2
maximum-flow is a skill published in the GitHub repository Arcadi4/nerdy (7 stars, last pushed 4mo ago), licensed MIT. It adds 59 tokens to every session and 3,824 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 2026-08-31.
Other skills, from other repositories
new-pattern
Guided workflow to create a new pattern following the project template and quality standards. Walks through topic validation, source verification, implementation, exercises, challenge questions, and bilingual docs.
coding-frameworks
Reference frameworks for coding interviews — the UMPIRE method, a pattern taxonomy (two-pointers, sliding-window, BFS/DFS, backtracking, dynamic-programming, graphs, and more), a Big-O cheatsheet, Python idioms, communication guidance, and anti-patterns. Used by the coding commands (/coding-explain, /practice-coding…
behavioral-frameworks
Reference frameworks (STAR, SBI, CARL), anti-patterns, and standard interviewer follow-ups for behavioral interview prep. Use when extracting STAR stories from user's experience, mapping stories to a specific company or JD, running mock behavioral interviews, rehearsing behavioral story delivery, or debriefing a real…
algo-sensei
Your personal DSA & LeetCode mentor. Use for problem explanations, progressive hints, code reviews, mock interviews, pattern recognition, complexity analysis, and custom problem generation. Automatically adapts to your learning style and request type.
campus-dsa-visualizer
Activate when a student, TA, or instructor asks to visualize the execution of a data structure or algorithm rather than just read its code — trigger phrasings include "visualize this binary search tree", "show me how quicksort partitions this array step by step", "trace this BFS/DFS on the whiteboard", "draw the DP…
nerd
Explains any code, system, or flow as ASCII flowcharts that expose the fundamental data structures and algorithms underneath. Use /skill:nerd to get a bird's-eye-view map of how something works, then zoom into any specific algorithm or data structure for a deeper breakdown. Strips jargon, shows the actual machine …