LLM Digest
Subscribe

Agent Builder Foundations

Mechanisms, math intuition, evidence, and application

View as JSON

Agent Builder Foundations

Mechanisms behind reliable agents

Durable explanations for builders who want to understand why a prompt, retrieval setup, tool call, memory system, or eval behaves the way it does. Each page separates paper-backed mechanisms, benchmark results, field reports, and LLM Digest synthesis.

24 concepts·7 clusters·evidence-tiered

Prompting and instruction following

Retrieval and grounding

Tool use and agents

Memory and context

  • Why does adding more context sometimes hurt an agent?

    Most production agent failures trace back to unmanaged context, not weak reasoning — treating context as a lifecycle to architect, ingest, scope, anticipate, and compact (not a log to truncate when it gets too big) is…

    7 evidence tiers
  • Why does CLAUDE.md (or AGENTS.md) only ever grow, never shrink?

    Agent instruction files grow because appending a rule is cheap while proving a rule is safe to delete becomes combinatorial once its rationale is forgotten — a 1,867-repository study found these files more than…

    4 evidence tiers
  • Can you trust what your agent remembers?

    Persistent agent memory is a write-once, replay-many attack surface — 2026 benchmarks show attackers can forge an agent's own reasoning history or plant poisoned facts through routine content like email, both with high…

    4 evidence tiers

Evals and reliability

  • What should an agent eval actually measure?

    An agent eval only earns its keep if it grades the trajectory (not just the final text), separates cheap deterministic graders from expensive model-based ones, and gets audited as hard as the agent — Anthropic's own…

    3 evidence tiers
  • Does adding memory to an agent actually make it better?

    Three independent 2026 evaluations agree that agent memory is not a universal win: the same technique gains one model 16 points of task completion, gains another zero, and most published memory frameworks actually…

    4 evidence tiers
  • Does a high benchmark score predict production reliability?

    A benchmark pass rate measures one round of scoring against a fixed task set — 2026 evidence shows agent-optimization gains that look real on that single round can fail to transfer or even regress once the agent is…

    11 evidence tiers
  • Can you trust an LLM-as-judge score?

    An LLM judge is a measurement instrument with its own biases, not ground truth — validate it the same way you validate the agent it grades, and for agent trajectories with checkable evidence, consider a deterministic…

    6 evidence tiers
  • Can you evaluate an agent's model router by replaying logged trajectories?

    No — a controlled branching-rollout study forked live SWE-bench agent trajectories at a model swap and found 61-94% of the actions after the swap diverge from what was logged, leaving only 3% of replayed states valid,…

    3 evidence tiers

Cost, latency, and operations

Safety and control