From Dev.to Community
Tech Articles
Curated developer articles, tutorials, and guides — auto-updated hourly
Mukunda Rao KattaMay 25, 2026 • 5 min read
prompt-shield: a tiny, zero-dep prompt-injection detector you can drop in front of any agent
A pattern-based first line of defense for LLM apps. Five rules, zero runtime deps, deterministic, fa...
#hermeschallenge#ai#llm#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Per-Step JSONL Logging for Agent Runs: Know What Your Agent Did and When
You shipped an agent to production. A user filed a bug report. The agent gave a wrong answer. You...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Make Your Agent Idempotent: Run It Twice, Get the Same Result
The user submitted the same form twice. Both requests hit your agent. The agent created two orders,....
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
The Full Conversation Lifecycle: From First Message to Stored Memory
How agent-message-window, conversation-codec, and agent-resume work together to manage a conversatio...
#hermeschallenge#ai#python#agents
1 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
Stop Writing One-Off Case Converters for LLM Tool Args: tool-arg-rename
Convert kwarg naming conventions for LLM tool calls with a single decorator, handling snake_case, ca...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
Format Tool Output for LLMs Before It Becomes a Tool Result: tool-output-format
Render tool output as LLM-friendly Markdown before returning it as a tool result, so the model reads...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
agenttrace-rs: Group LLM Calls into Named Runs and Get Cost Breakdowns in Rust
A Rust crate that records LLM calls into named runs and reports total cost, p50/p95 latency, and per...
#hermeschallenge#ai#rust#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Single-Flight LLM Calls: Coalesce 50 Concurrent Identical Requests Into One API Call
Fifty users submitted their daily summaries at 9:00 AM. All fifty requests ask the same follow-up...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Per-Tool Call-Count Caps: Prevent Any Single Tool From Running Away
Your agent has a web_search tool. During a long task, the model calls it 47 times. Your search API.....
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
Debug Agent Failures After They Happen with JSONL Traces
Using agenttap and agent-replay-trace to reconstruct and step through agent runs for post-incident d...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 6 min read
Production Python Agent Checklist: 15 Things Before You Deploy
15 things to check before deploying a Python LLM agent to production. Each one has a library or patt...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
agentsnap: Jest-Style Snapshot Tests for AI Agent Tool Calls
A JavaScript/TypeScript library that captures and diffs agent tool-call traces like Jest snapshots. ...
#hermeschallenge#ai#javascript#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
My RAG system slowly got worse for three months and nobody noticed.
A composable drift detector for RAG and agent systems. Five orthogonal dimensions, one DriftReport, ...
#hermeschallenge#ai#rag#python
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
Your Agent Loop Has No Clock: agent-deadline
A cooperative per-task deadline for agent loops that raises DeadlineExceeded at the next check_or_ra...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Per-Key Rate Limiting for Agent Tool Calls: Stop One User From Breaking Everything
Multi-tenant agents share infrastructure. When one user's agent calls the web search tool 200 times....
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
The $12 LLM Call Nobody Saw Coming: llm-cost-cap
A zero-dependency Python library that estimates input token cost before an LLM call and raises CostC...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
Stop Your Agent from Sneaking in 'I would recommend...' When You Said Only 'escalate', 'resolve', or 'defer'
llm-output-validator is a rule-based Python library that validates LLM string outputs against compos...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 6 min read
Why Every Library in the Agent Stack Has Zero Runtime Dependencies
The zero-deps design choice for composable agent libraries: why it matters and what it costs.
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
My agent cited a document that does not exist. Here is the library I wrote to catch that.
agent-citation is a Python library for structured WHERE-layer citations in AI agent outputs, with pu...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Session-Scoped Tool Call Deduplication: Stop Your Agent From Doing the Same Thing Twice
The agent is looping. It called search_documents(query="Q3 revenue") in step 2. Now it is step 7 and...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
tool-arg-fuzzy-rs: Fuzzy Match LLM Enum Args to Valid Values in Rust
A Rust crate that cascades exact, case-insensitive, prefix, and substring matches to recover from LL...
#hermeschallenge#ai#rust#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
llm-budget-window: Per-Minute and Per-Hour Token Caps That Actually Work
A Rust crate with sliding time windows for token and USD budgets. Blocks before the expensive call, ...
#hermeschallenge#ai#rust#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 4 min read
Durable Agent State Checkpoints: Resume Where You Left Off When the Run Crashes
Your agent runs for 4 minutes. At minute 3, the process gets OOM-killed. Or the network drops. Or th...
#hermeschallenge#ai#python#agents
0 0
Mukunda Rao KattaMay 25, 2026 • 5 min read
llm-circuit-breaker-py: Open the Circuit Before Your Agent Hammers a Down Provider
A Python port of llm-circuit-breaker with Closed/Open/HalfOpen states, thread-safe, sync and async s...
#hermeschallenge#ai#python#agents
0 0