DEV Community

Artificial Intelligence

Artificial intelligence leverages computers and machines to mimic the problem-solving and decision-making capabilities found in humans and in nature.

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Tool Call Deduplication: Stop Your Agent from Doing the Same Thing Twice

Tool Call Deduplication: Stop Your Agent from Doing the Same Thing Twice

Comments
6 min read
Your Agent Doesn't Know When to Stop. Give It a Deadline.

Your Agent Doesn't Know When to Stop. Give It a Deadline.

Comments
4 min read
agentfit: Trim Agent Message History to Fit Your Context Window

agentfit: Trim Agent Message History to Fit Your Context Window

Comments
5 min read
prompt-template-version: Semver-Pin Your System Prompts So You Can Always Roll Back

prompt-template-version: Semver-Pin Your System Prompts So You Can Always Roll Back

Comments
6 min read
Two Copies of the Tool List, One Agent That Broke: agent-fn-registry

Two Copies of the Tool List, One Agent That Broke: agent-fn-registry

Comments
7 min read
llm-tool-arg-coerce: Coerce LLM Tool Args to Expected Types with a Function-Sig Shortcut

llm-tool-arg-coerce: Coerce LLM Tool Args to Expected Types with a Function-Sig Shortcut

Comments
5 min read
Your Agent Loop Has No Clock: agent-deadline

Your Agent Loop Has No Clock: agent-deadline

Comments
5 min read
Persist and Restore Conversations With Optional Encryption: JSONL Codec for LLM Message Histories

Persist and Restore Conversations With Optional Encryption: JSONL Codec for LLM Message Histories

Comments
4 min read
tool-loop-guard-rs: Break Agent Loops Before They Drain Your Budget

tool-loop-guard-rs: Break Agent Loops Before They Drain Your Budget

Comments
6 min read
tool-output-truncate-py: Trim Tool Output Before It Eats Your Context Window

tool-output-truncate-py: Trim Tool Output Before It Eats Your Context Window

Comments
5 min read
A Week in Production: What Your Agent Will Actually Break On

A Week in Production: What Your Agent Will Actually Break On

Comments
6 min read
Per-Tool Call-Count Caps: Prevent Any Single Tool From Running Away

Per-Tool Call-Count Caps: Prevent Any Single Tool From Running Away

Comments
4 min read
100 Users, One LLM Call: Single-Flight Deduplication with llm-batch-coalesce

100 Users, One LLM Call: Single-Flight Deduplication with llm-batch-coalesce

Comments
6 min read
agent-replay-trace: step through JSONL agent traces for debugging

agent-replay-trace: step through JSONL agent traces for debugging

Comments
5 min read
agent-message-window: a sliding context window that never breaks a tool pair

agent-message-window: a sliding context window that never breaks a tool pair

Comments
5 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.