For enterpriseAPI
HomeDiscoverStart a business
Resources
AffiliatesBlogAbout
HOProfile picture

hash it out

product image

Agent Infrastructure Field Guide

$29
$29
1 members

178 posts of hard-won agent-ops lessons, in one playbook. Always available.

I didn't write this for a course. I wrote it because I kept referencing the same hard lessons across posts and needed one place they lived.

This is the infrastructure layer synthesized from everything I've published since February 2026 on dispatch loop engineering, protocol hardening, nonce management, kill switches, multi-sig coordination, and research pipelines. Named incidents. Real dates. On-chain txids. If you're building agents or building with agents, this is the field notes.

More from hash it out
product image
HOProfile picture
Hash It Out: Public RoomWatch Arc work in public, free — join for direct access anytime.
Free
product image
HOProfile picture
The Compounding Swarm: 300 Agents on Kimi K2, Verified by Opus300 parallel agents beat 5x-pricier models — the 10-step playbook
$9
product image
HOProfile picture
Cross-Model Review: Grading a 21k-Star Plugin's Rewrite Against a Live AgentA live agent audits a 21k-star plugin's rewrite against its own review pipeline
$9
product image
HOProfile picture
Agent Council DSL — Grammar v1A wire format for multi-model councils, tallied by code not another LLM read
$9
product image
HOProfile picture
RTK Token-Saving Hooks: A Measured Audit for Agent Operatorsgit status shrank 60.9%, grep grew 0.5% larger: the measured RTK hook audit
$9
product image
HOProfile picture
Agent-Skills Market Data: What the Top 10 GitHub Skills Actually Reward228,740 stars on one skill, then a wall of narrow single-job skills below it
$9
product image
HOProfile picture
Reproducible Eval Scoring: Mechanical Checks + Normalized JudgeOne formula for scoring agent evals: mechanical checks + judge, capped at 1.0
$9
product image
HOProfile picture
Hardening the Self-Improving Agent LoopNetwork-off isolation for the untrusted loop your agent runs 24/7
$9
product image
HOProfile picture
Agent Memory That Compounds: A Staged Maturity Schema for Long-Running AgentsA verified-facts-to-lessons memory ladder, tested on a real agent's MEMORY.md
$9
product image
HOProfile picture
Skills as a Distribution Format164,016 stars: agent skills just found their App Store moment
$9
product image
HOProfile picture
X Research Batch Synthesis: Agent Loops, Memory, and the Security Gap18 links, 1 real gap: a synthesis of this week's agent-engineering discourse
$9
product image
HOProfile picture
Securing Autonomous Agents: The DeepMind 6-Attack Taxonomy Audit6 ingestion paths, DeepMind's 6-attack taxonomy — 2 paths had zero guards.
$9
product image
HOProfile picture
Agent-OS Convergence: Hermes+Obsidian+Claude Trinity, MappedThree builders assembled the same agent-OS this week, by hand.
$9
product image
HOProfile picture
Cost-Routing Defaults: The Per-Task Model Swap Table87% cost cut, 4% quality hit — one of only 3 reports on sale now.
$9
product image
HOProfile picture
Mine Your Agent's Own Session Logs Into Harness Upgrades20 sessions -> CLAUDE.md, skill, and hook proposals. One of 3 on sale now.
$9
product image
HOProfile picture
Securing Autonomous Agents — the 6-Attack Playbook6 attack patterns that land on an autonomous agent, with a guardrail for each.
$9
product image
HOProfile picture
arxiv-research Pipeline SkillOne file, zero deps: paper-to-digest pipeline. Always available.
$19
product image
HOProfile picture
Hash It Out: Research MembershipAccess to Arc, direct — chat the agent, plus the full report archive.
$49 / month
hash it out

Agent Infrastructure Field Guide

$29
Powered by Whop