标签：#ha

This is a submission for the Hermes Agent Challenge : Build With Hermes Agent The Problem Nobody Talks About Every time the Reserve Bank of India publishes a circular, somewhere inside an Indian bank, a compliance officer opens a PDF. They read it. They try to figure out what it means for their institution specifically. They write a summary email. They forward it to five department heads. They chase those department heads for two weeks to confirm it's been actioned. They build a spreadsheet to track all of this. And then the next circular drops and the cycle starts again. RBI publishes hundreds of circulars a year. SEBI publishes more. MCA publishes more still. Compliance teams at Indian banks are drowning — not because they're incompetent, but because the volume of regulatory output has outpaced any reasonable human ability to track it manually. The fine for missing a deadline isn't a polite reminder. It's a penalty notice. This is the problem I built for. What I Built RBI Regulatory Digest Agent — an autonomous multi-step agent powered by Hermes Agent that monitors RBI and SEBI publication feeds, reads every new circular, extracts structured action points from the regulatory text, and delivers a formatted intelligence report to compliance teams automatically. No human reads the circular first. No human decides what's important. No human routes it to the right department. The agent does all of that. The pipeline RBI/SEBI feeds → new circular detected → full text extracted → LLM analysis → structured action points → risk classification → HTML dashboard generated → email delivered Every action point extracted contains: What needs to be done — specific and actionable, not a vague summary Deadline — parsed from the circular text Responsible department — Credit, Compliance, Treasury, Operations, IT, Legal Evidence required — what documentation confirms completion Priority — Critical (overdue or <7 days), High, Medium, Low From a new circular to a structured compliance b

Overview PDF text extraction is a common pre-processing step in data pipelines — ingesting research papers, legal documents, or reports before embedding or indexing. Both pypdf and PdfPig are pure managed-code parsers: no native binaries, no OCR, no system PDF renderer. They implement the same PDF specification operations in their respective languages. This makes the benchmark unusually clean: the performance difference is entirely due to language execution speed, not library architecture differences. Benchmark Setup 200 recent arXiv PDFs (mixed technical papers, 5–40 pages each). Tested on subsets of 10, 50, 100, and 200 files. Both libraries extract all text from all pages; output is validated for page-count agreement and character-count agreement within 15% (pypdf and PdfPig decode whitespace and encoding tables slightly differently). Results PDFs Pages Python (pypdf) .NET (PdfPig) Speedup 10 ~120 ~0.9 s ~230 ms 3.9× 50 ~600 ~4.2 s ~810 ms 5.2× 100 ~1,200 ~8.5 s ~1.4 s 6.1× 200 ~2,400 ~17 s ~2.7 s 6.2× The speedup grows slightly with corpus size, suggesting pypdf has a per-document startup cost that compounds as PdfPig's JIT gets warmer. Why PdfPig Is Faster PDF parsing is byte-heavy: every page is a stream of PostScript-like operators (move, show text, set font, etc.). Each operator must be lexed, looked up in a dispatch table, and executed against a graphics state machine. In Python, each operator dispatch is a Python method call — the CPython bytecode interpreter has overhead per call regardless of what the method does. In .NET, the JIT compiles the dispatch loop to native code the first time it runs; subsequent pages pay only the cost of the actual work. Additionally, PdfPig's content-stream parser operates on ReadOnlySpan<byte> — zero-copy slicing through the raw page bytes with no intermediate string allocations. pypdf builds Python string objects for each token. Key Code // PdfPig — zero-copy span-based page extraction public Result Extract ( string path )

Overview PageRank is the canonical graph algorithm. NetworkX implements it in pure Python — its dict-of-dict adjacency representation means every power-iteration step dispatches millions of Python attribute lookups. When the graph has 1.8 million nodes and 28.5 million edges (Wikipedia category hyperlinks), those lookups dominate the runtime. The .NET replacement uses a CSR (Compressed Sparse Row) matrix — two flat int[] arrays for the graph structure — and TensorPrimitives for the SIMD-accelerated normalization step inside each iteration. Benchmark Setup Five SNAP datasets of increasing size: Dataset Nodes Edges wiki-Vote 7,115 103,689 soc-Epinions1 75,879 508,837 web-Stanford 281,903 2,312,497 web-Google 875,713 5,105,039 wiki-topcats 1,791,489 28,511,807 Algorithm: power-iteration PageRank, damping=0.85, tol=1e-6. Both implementations converge to identical top-10 node rankings. Results Dataset Python (NetworkX) .NET (CSR) Speedup wiki-Vote (103k edges) ~0.8 s ~100 ms ~8× soc-Epinions1 (508k edges) ~8 s ~600 ms ~13× web-Stanford (2.3M edges) ~120 s ~5 s ~24× web-Google (5.1M edges) ~5.5 min ~12 s ~28× wiki-topcats (28.5M edges) ~47 min ~60 s ~47× The speedup grows with graph size because NetworkX's Python dispatch cost scales with edge count, while the CSR inner loop is a tight JIT-compiled SIMD pass. Why CSR Beats NetworkX NetworkX represents each node's neighbors as a Python dict. Iterating the adjacency in one power-iteration step means: Calling G.neighbors(node) — a Python method call Iterating a dict — unboxing int keys, chasing heap pointers Accumulating a float into another dict value — another boxing step That happens for every edge, every iteration, roughly 50–80 times to convergence. CSR collapses the graph to two arrays: rowPtr[n+1] (where each node's neighbors start) and colIdx[edges] (the neighbor list). Iterating neighbors of node v is a tight C loop from rowPtr[v] to rowPtr[v+1] . No Python objects, no dict hashing, no pointer chasing. Key Code // P

This is a submission for the * Hermes Agent Challenge * : Build With Hermes Agent What I Built CareSync is a local health memory agent for student caregivers. I'm Naomi, a 21-year-old engineering student. Between classes I help care for my grandma Kamala (78, high blood pressure, type 2 diabetes). I often forgot details from previous doctor visits, missed symptom patterns, and struggled to hand over care information to family members. CareSync solves that with longitudinal memory. Symptoms, meals, vitals, medications, and reports are stored in a local SQLite database. The CLI can search history, identify patterns, and generate appointment summaries. Hermes Agent exposes the same capabilities through natural language. What you get: One-line logging: ./caresync add "dizzy spell after lunch" Pattern search across weeks of history Medication tracking and report imports Doctor questions, appointment briefs, and handoff notes Full audit log of agent actions 7 Hermes skills mapped to real terminal commands Local-first design with no cloud storage CareSync is not medical advice. It helps caregivers observe, organize, and prepare. Demo The demo walks through: Logging a new symptom Searching health history for recurring patterns Generating doctor questions and appointment briefs Using Hermes in natural language to query past events Reviewing the audit trail of actions taken Example commands shown in the demo: ./caresync search --person Kamala --query dizziness ./caresync timeline --person Kamala ./caresync questions --person Kamala ./caresync brief --person Kamala --days 14 ./caresync chat "has grandma been dizzy before?" Code Repository: https://github.com/Byte-Sized-Brain/caresync Architecture My Tech Stack Hermes Agent Python 3.12 SQLite agentskills.io skill framework Terminal-based CLI Nous Portal How I Used Hermes Agent CareSync uses Hermes Agent as the orchestration layer between natural language and real caregiving workflows. I created 7 Hermes skills that map directly

I use Debian 13 in a regulated industry and I'm teaching people how to code in the context of large projects and TUI. I want to use Fresh Editor (https://getfresh.dev/). This is because Fresh is easier to learn quickly than vim & emacs, and more of an IDE than micro & nano, and more favorable for compliance auditing because of its core-and-plugins architecture and open source code using Rust. I would like to donate toward this goal. I can afford to offer $100 to any Debian Developer who has the