Agent Mentor Learn

Glossary

66 terms from “Verification and Quality Assurance: Don't Let 'Looks Right' Slip Through.” Hover the first occurrence in the lesson for its definition.

TermDefinitionSource
looks doneClaude stops when the work looks done; without a check it can run, 'looks done' is the only signal available in the entire system.Best practices for Claude Code — Claude Code official documentation
verification loopThe role that falls to you when there's no runnable check in the pipeline: every mistake waits for you to notice it.Best practices for Claude Code — Claude Code official documentation
trust-then-verify gapThe official name for this phenomenon: Claude produces a plausible-looking implementation that doesn't handle edge cases; you trust it first, and verification either doesn't happen or happens too late.Best practices for Claude Code — Claude Code official documentation
assertionStatements you can only choose to believe or not: 'logic is correct,' 'should be fine,' 'already optimized,' 'no errors during execution.'Best practices for Claude Code — Claude Code official documentation
evidenceSomething a second person can re-run exactly the same way: a command plus its raw output, an exit code, a list of failed test names, a screenshot, a before/after numeric comparison.Best practices for Claude Code — Claude Code official documentation
deterministic systemIn computing, systems that produce the same output every time given identical inputs; this course's verifiers are this kind of 'predictably dumb' thing, using a deterministic thing to gate a non-deterministic thing.Writing effective tools for agents — with agents — Anthropic Engineering
non-deterministicAgent systems can generate varied responses even with identical starting conditions; even if the prompt hasn't changed a single character, decisions across runs aren't guaranteed to match.Writing effective tools for agents — with agents — Anthropic Engineering
compounding errorsIn agent systems errors snowball: one step failing causes agents to explore entirely different trajectories, and they make new decisions based on bad results, outcomes become unpredictable.How we built our multi-agent research system — Anthropic Engineering
sandbox environmentWhere Anthropic recommends agents be thoroughly tested: autonomy means higher costs and potential for compounding errors, so test in sandboxes with appropriate guardrails.Building Effective AI Agents — Anthropic Engineering
ground truthReal feedback agents get from the environment during execution, like tool call results or code execution results, used to assess their own progress.Building Effective AI Agents — Anthropic Engineering
demonstrably improves outcomesThe admission condition for adding complexity: you should consider adding complexity only when it demonstrably improves outcomes.Building Effective AI Agents — Anthropic Engineering
stop_reasonThe field in model responses that drives the harness loop: when the value is tool_use, continue executing tools and feeding back; when it becomes end_turn, the loop exits.Tool use with Claude — Claude API documentation
end_turnOne value of stop_reason, meaning the model doesn't plan to call more tools this turn, nothing more.Tool use with Claude — Claude API documentation
tool_resultThe message block that feeds tool execution results back to the model, paired with the corresponding tool_use via tool_use_id; verifier output flows back into the conversation through this.Tool use with Claude — Claude API documentation
run_checkWrapping a verification script as a tool the model can call, letting it run the check itself within the loop and read results, with exit code and stdout passed through as-is.Best practices for Claude Code — Claude Code official documentation
exit codeHow verification scripts express conclusions: 0 means pass, non-zero means fail; CI and shell && can use it directly.Best practices for Claude Code — Claude Code official documentation
pass/failThe objective binary result a check can produce; with it the loop closes on its own—Claude does work, runs check, reads result, iterates until check passes.Best practices for Claude Code — Claude Code official documentation
fixtureA golden-answer file saved beforehand; after running, compare output to it; the script in the official check menu that 'diffs output against it' uses this.Best practices for Claude Code — Claude Code official documentation
end stateThe default judgment method: don't judge whether the agent followed a specific process, judge whether it achieved the correct final state.How we built our multi-agent research system — Anthropic Engineering
turn-by-turn analysisThe approach end-state evaluation replaces: checking agent actions turn by turn, attempting to validate every intermediate step.How we built our multi-agent research system — Anthropic Engineering
verification checkpointA few discrete observation positions in complex workflows, where you verify 'specific state changes should have occurred' rather than validating every intermediate step.How we built our multi-agent research system — Anthropic Engineering
recovery checkpointThe Course 9 meaning: writing the agent's running state to disk so it can resume from there after a crash; purpose is recovery, not judgment.How we built our multi-agent research system — Anthropic Engineering
success criteriaThe set of requirements that turn 'correct end state' into concrete numbers or clear judgments; if you can't write it no matter what, this task shouldn't have been handed entirely to the agent to run alone.Define success criteria and build evaluations — Claude API documentation
measurableThe first hard requirement for success criteria: use quantitative metrics or well-defined qualitative scales; numbers provide clarity and scalability.Define success criteria and build evaluations — Claude API documentation
achievableThe second hard requirement for success criteria: base targets on industry benchmarks, prior experiments, AI research, or expert knowledge; shouldn't be unrealistic to current frontier model capabilities.Define success criteria and build evaluations — Claude API documentation
multidimensional evaluationMost use cases need evaluation along several success criteria—dimensions undermine each other, leaving less room for shortcuts.Define success criteria and build evaluations — Claude API documentation
trajectory assertionAn optional addition: for a set of prompts and responses, additionally specify which tools you expect the agent to call, to measure whether it grasps each tool's purpose.Writing effective tools for agents — with agents — Anthropic Engineering
expectedToolsThe optional field in eval cases that carries trajectory assertion: expects it touched at least these tools, regardless of order or call count.Writing effective tools for agents — with agents — Anthropic Engineering
redundant tool callsOne diagnostic metric: call count is noticeably high, usually suggests pagination or token limit parameters need rightsizing.Writing effective tools for agents — with agents — Anthropic Engineering
tool errorsOne diagnostic metric: lots of errors due to invalid parameters usually suggests these tools need clearer descriptions or better examples.Writing effective tools for agents — with agents — Anthropic Engineering
code-based gradingFirst place in the grading method ranking: fastest, most reliable, extremely scalable; the weakness is it lacks nuance for complex judgments requiring less rule-based rigidity.Define success criteria and build evaluations — Claude API documentation
LLM-based gradingSecond place in the ranking: fast, flexible, scalable, suitable for complex judgment; prerequisite is test to ensure reliability first, then scale.Define success criteria and build evaluations — Claude API documentation
human gradingThird place in the ranking: most flexible, highest quality, but slow and expensive; avoid if possible.Define success criteria and build evaluations — Claude API documentation
exact matchThe leftmost form on the verifier spectrum: measures whether model output matches a predefined correct answer, typically after normalizing whitespace and case.Define success criteria and build evaluations — Claude API documentation
output == golden_answerThe minimal form of exact match, just an equality check; simple, unambiguous, suitable for tasks with clear-cut categorical answers.Define success criteria and build evaluations — Claude API documentation
normalizationBefore comparing, erase differences that don't carry meaning: fold consecutive whitespace, trim leading/trailing, unify case; for amounts, also wash away currency symbols, thousand separators, units.Define success criteria and build evaluations — Claude API documentation
deterministic verifierThis course's core technique: using code that gives the same conclusion every time to gate a non-deterministic system's output, returning pass/fail and readable failure reasons.Writing effective tools for agents — with agents — Anthropic Engineering
spectrumVerifiers aren't a few mutually exclusive options but a continuous band: one end is exact string comparison against baseline, the other end is asking Claude to judge.Writing effective tools for agents — with agents — Anthropic Engineering
strict: trueA line added to tool definitions, ensuring Claude's tool calls strictly conform to your declared schema, moving one class of structure validation forward as a platform-level guarantee.Tool use with Claude — Claude API documentation
false negativeThe verifier judges correct output as failure: it doesn't let errors through, it wrongly accuses correct ones.Writing effective tools for agents — with agents — Anthropic Engineering
overly strict verifierThe most common way deterministic checks fail: due to spurious differences like formatting, punctuation, or valid alternative phrasings, they judge correct responses as failures.Writing effective tools for agents — with agents — Anthropic Engineering
free textOutputs like research and summaries: free-form, rarely have a single correct answer, difficult to evaluate programmatically; LLMs are a natural fit for scoring this output.How we built our multi-agent research system — Anthropic Engineering
human reviewA step still indispensable beyond automated testing: automated tests verify functionality is correct, but ensuring solutions align with broader system requirements still needs human eyes.Building Effective AI Agents — Anthropic Engineering
LLM judgeUsing a model call to score free-form text output, positioned where deterministic checks can't reach, not as their replacement.How we built our multi-agent research system — Anthropic Engineering
rubricBreaking down vague 'is it good' into several concrete questions, answering and scoring each separately; for research tasks the ready-made breakdown is five dimensions.How we built our multi-agent research system — Anthropic Engineering
reason first then scoreA key technique for judge prompts: have it write reasoning first then produce a score, discard reasoning afterward; improves grading performance, especially noticeable for complex judgments.Define success criteria and build evaluations — Claude API documentation
fresh contextWhere the reviewer should sit: only sees output and the criteria you gave, can't see the reasoning process that produced this change, so judges the result on its own terms.Best practices for Claude Code — Claude Code official documentation
self-reportThe agent's description of its own process ('I searched three authoritative sources, cross-checked and confirmed the percentage'); doesn't count as evidence by itself.Writing effective tools for agents — with agents — Anthropic Engineering
raw transcriptThe execution log including tool calls and tool responses, used to catch behaviors not explicitly stated in the agent's chain-of-thought.Writing effective tools for agents — with agents — Anthropic Engineering
find problemsThe judge's first failure mode: a reviewer prompted to find gaps will usually report some, even when the work is sound—because that's what you asked it to do.Best practices for Claude Code — Claude Code official documentation
over-engineeringThe consequence of chasing every review finding: extra abstraction layers, defensive code, and tests for cases that can't possibly happen.Best practices for Claude Code — Claude Code official documentation
effect sizeHow big the gap is from one change—is it from 71.2% to 72.4%, or from 30% to 80%; larger gaps need fewer samples.How we built our multi-agent research system — Anthropic Engineering
eval setA batch of tasks paired with verifiable results; official starting scale is about 20 queries representing real usage; don't wait to hoard hundreds.How we built our multi-agent research system — Anthropic Engineering
verifiable resultWhat every eval prompt should be paired with: a judgeable end state, string, state change, or rubric; without it the prompt isn't an eval task, it's a demo.Writing effective tools for agents — with agents — Anthropic Engineering
edge caseLow-frequency but actually-occurring situations (missing info, mixed demands, tool errors); the insurance in eval sets, not the main body.Define success criteria and build evaluations — Claude API documentation
volume over qualityAn eval set design principle: more questions with slightly lower signal automated grading beats fewer questions with high-quality human hand-grading.Define success criteria and build evaluations — Claude API documentation
real distributionWhat evals should stick to: your eval tasks should be grounded in real-world usage, able to point to any prompt and say 'three users asked this way last week.'Define success criteria and build evaluations — Claude API documentation
held-out setA batch of cases separated from the start, not looked at or run or touched during daily tuning, relies on it to ensure no overfitting to the 'training' evals.Writing effective tools for agents — with agents — Anthropic Engineering
dev setThe batch of cases run repeatedly when changing prompts daily, the more frequently the better; its scores are navigation, not conclusion.Writing effective tools for agents — with agents — Anthropic Engineering
overfittingThe entire set of prompt plus tool configuration learned features of the eval materials rather than patterns of the task itself—like hardcoding a sentence in system prompt for one persistently failing case.Writing effective tools for agents — with agents — Anthropic Engineering
not guaranteedOfficial docs' wording for certain ideal behaviors (like the model will proactively ask for missing required parameters); especially so for more ambiguous prompts and less capable models.Tool use with Claude — Claude API documentation
eval trackThe runnable program that welds eval sets, layered grading, and harness loops together: type one command, scores tell you which got better, which didn't move, whether anything that was right got broken.Writing effective tools for agents — with agents — Anthropic Engineering
stub clientA fake messages.create that spits responses in a fixed queue order, turning model behavior into a controlled variable, making the whole track reproducible.Writing effective tools for agents — with agents — Anthropic Engineering
response queueThe pre-written response sequence behind the stub client; the difference between two prompt versions is pinned by two sets of queues, 'assuming new prompt takes effect and model answers this way' is encoded as data.Writing effective tools for agents — with agents — Anthropic Engineering
task isolationOne eval task one independent loop, one independent messages; finished then discarded; implementation is just don't hoist messages outside the function.Writing effective tools for agents — with agents — Anthropic Engineering
is_errorThe marker attached to tool_result when feeding back tool exceptions: catch the exception, wrap it into a result with this marker to return to the model, simultaneously incrementing the error counter.Tool use with Claude — Claude API documentation