Domain 5: Context Management & Reliability (15%)
6 lessons, each ending with a practice scenario (Pro)
Log in to save your progress.
- ✓5.1 — Context Window ManagementThe context window is everything Claude reads on one request. Managing it is choosing what goes in.Scenario: Pro
- ✓5.2 — Escalation & Ambiguity ResolutionWhen a tool returns multiple matches: always ask for additional identifiers. Never pick by "most recent" or "most active" heuristic (privacy risk).Scenario: Pro
- ✓5.3 — Error Propagation in Multi-Agent SystemsMiddle ground: structured error propagation → coordinator decides recovery. Access Failure vs Valid Empty Result: Meaning, Retry?Scenario: Pro
- ✓5.4 — Codebase Exploration & Context DegradationContext degradation = model loses grip on earlier specific findings (refers to "typical patterns" instead of class names). It is an attention quality problem, not a token-limit problem. Larger context windows do not fix it.Scenario: Pro
- ✓5.5 — Human Review & Confidence Calibration97% overall accuracy can hide catastrophic failure rates on specific document types or fields (e.g. handwritten receipts at 60%). Always validate by document type and field segment before automating.Scenario: Pro
- ✓5.6 — Information Provenance & Multi-Source SynthesisEvery finding must include:; Claim; Source URL; Document name; Relevant excerpt; Publication date.Scenario: Pro
Domain recapQuick overview of all lessons
6 Task Statements / Lessons
5.1 Context Window Management
Strategies when the window fills: summarisation, sliding window, subagent isolation, progressive disclosure, prompt caching of static content.
5.2 Escalation & Ambiguity Resolution
Detecting ambiguity, asking clarifying questions, or escalating to a human with the right context package.
5.3 Error Propagation in Multi-Agent Systems
How failures in one agent affect downstream agents; circuit breakers, partial-result handling, and recovery.
5.4 Codebase Exploration & Context Degradation
Techniques for exploring large codebases without drowning the context window; when context quality degrades and how to detect it.
5.5 Human Review & Confidence Calibration
When to surface confidence scores or force human review; calibrating the model’s expressed confidence.
5.6 Information Provenance & Multi-Source Synthesis
Tracking where information came from, reconciling conflicting sources, citing or attributing correctly.
Practical tip: Prompt caching of the system prompt + static knowledge is almost always the first optimisation for cost and latency.