| No. | Mind map for | Ask yourself | Objective |
|---|
| Domain 1: Agentic architecture and orchestration |
| 1 | Fixed sequence or model-driven loop | Is the path known in advance, or does each step depend on what was just found? | 1.1 Agentic loops |
| 2 | When an agent loop should continue or stop | Which signal is the loop reading to decide it is done? | 1.1 Agentic loops |
| 3 | Why an agent repeats calls or forgets what it found | What is missing from the history the loop sends back to the model? | 1.1 Agentic loops |
| 4 | Why Claude talks about stale or out-of-date code | Is Claude reasoning from an old tool result, or from a context that has degraded? | 1.1 Agentic loops |
| 5 | Finding where a research report's gap started | When could the missing area first have been seen: in the request itself, or only after reading the sources? | 1.2 Coordinator and subagents |
| 6 | Cutting waste in a slow or costly multi-agent pipeline | Which waste is biggest: unneeded subagents, overlapping work, needless sequencing or real-time pricing? | 1.2 Coordinator and subagents |
| 7 | When one subagent needs another subagent's help | Can this request change the plan, or is it a simple, frequent need? | 1.2 Coordinator and subagents |
| 8 | Whether to hand work to a subagent at all | Would delegating isolate verbose output or run independent work in parallel? | 1.3 Subagent invocation and context passing |
| 9 | Where each piece of a subagent's context belongs | Does this information change per run, and where must it be written so the subagent actually sees it? | 1.3 Subagent invocation and context passing |
| 10 | Why a subagent's output is weak, generic or missing | Which step of delegation failed: spawning, choosing the subagent, briefing it, or the inputs it was given? | 1.3 Subagent invocation and context passing |
| 11 | Speeding up a slow multi-agent run | Is the time going on needless sequencing, unneeded subagents, or a real dependency? | 1.3 Subagent invocation and context passing |
| 12 | How to brief a coordinator agent | Do questions arrive in new shapes, or are they known, stable types? | 1.3 Subagent invocation and context passing |
| 13 | Prompt or code for a rule that must hold | What does one miss cost: is this a must-always rule or a should-usually one? | 1.4 Multi-step workflows, enforcement and hand-off |
| 14 | Which control enforces an order or forces a tool call | Do you need to block a wrong call, force one call on one turn, or enforce an order across a whole session? | 1.4 Multi-step workflows, enforcement and hand-off |
| 15 | Organising a request that raises several concerns | Are the concerns lost in the first reply or many turns later, and how heavy is each one? | 1.4 Multi-step workflows, enforcement and hand-off |
| 16 | Handing work over to a human or a later session | Who picks the work up next, and what do they need to read first? | 1.4 Multi-step workflows, enforcement and hand-off |
| 17 | Fixing a gate that blocks a legitimate case | Can the proof be seen in tool data, so the gate can learn a second kind of proof in code? | 1.4 Multi-step workflows, enforcement and hand-off |
| 18 | Hard stop, cleanup or nudge: which control acts at the right moment | Must the rule hold every time, and does it act on the outgoing call or on the result? | 1.5 Agent SDK hooks |
| 19 | Where to fix messy, sensitive or derived tool data | Do you control the source, and is the job deterministic enough for code? | 1.5 Agent SDK hooks |
| 20 | When a hook exists but the policy still breaks | What does the hook actually check, and does that match what the policy is about? | 1.5 Agent SDK hooks |
| 21 | Fixed chain or adaptive decomposition for a multi-step job | Can the steps be listed before the work starts, or does reading reveal new work? | 1.6 Task decomposition |
| 22 | Making agent runs faster or cheaper | Does each request need the full pipeline, and is anyone waiting for the answer? | 1.6 Task decomposition |
| 23 | Strategy for understanding unfamiliar code | Do you need the answer to one specific question, or a broad map of a large area? | 1.6 Task decomposition |
| 24 | Splitting one growing prompt | Is the problem where items sit in the prompt, or that one prompt can't fit every kind of input? | 1.6 Task decomposition |
| 25 | Resume, fork or start fresh when continuing old work | Is the earlier session still accurate, and do you want one line of work or two? | 1.7 Session state, resuming and forking |
| Domain 2: Tool design and MCP integration |
| 26 | Claude picks the wrong tool: which lever to pull first | Is the tool description the problem, or is something else steering the choice? | 2.1 Tool descriptions and boundaries |
| 27 | A tool is called with bad inputs: description, examples or a split | What was the model actually told about this tool and its inputs? | 2.1 Tool descriptions and boundaries |
| 28 | What error to return for each kind of tool failure | What does the agent need to know about this failure to recover from it? | 2.2 Structured tool errors |
| 29 | Telling "found nothing" apart from "could not look" | Did the tool look and find nothing, or could it not look at all? | 2.2 Structured tool errors |
| 30 | What a failing subagent sends back to the coordinator | Can this failure be fixed locally, and if not, what does the coordinator need in order to decide? | 2.2 Structured tool errors |
| 31 | Which tool_choice setting to use | What may Claude still choose: text or a tool, which tool, or nothing at all? | 2.3 Tool distribution and tool_choice |
| 32 | Too many tools, or the wrong tools, on one agent | Is the agent confused among the right tools, or holding tools it shouldn't have? | 2.3 Tool distribution and tool_choice |
| 33 | A subagent needs help from outside its role | What kind of outside help is it: a frequent simple check, or rare open-ended work? | 2.3 Tool distribution and tool_choice |
| 34 | Controlling a risky tool: remove it, gate it, or guide it | Is this action forbidden, allowed with approval, allowed under a limit, or just a preference? | 2.3 Tool distribution and tool_choice |
| 35 | Reuse a community MCP server, describe conventions, or build your own | Does something already exist for this system, and is the gap capability or just context? | 2.4 MCP servers in Claude Code and agents |
| 36 | Where an MCP server's configuration lives | Who needs this server, and how do they get it without a secret being committed? | 2.4 MCP servers in Claude Code and agents |
| 37 | A better MCP tool is ignored or missing | Is the MCP tool missing from Claude's list, or listed but losing to a built-in? | 2.4 MCP servers in Claude Code and agents |
| 38 | Too many exploratory calls: what the agent is missing | Is the agent calling tools to find out what exists, or because a tool is unclear? | 2.4 MCP servers in Claude Code and agents |
| 39 | Which built-in tool to find and explore code | Are you matching names, searching contents, reading a file, or only after a conclusion? | 2.5 Built-in tools (Read, Write, Edit, Bash, Grep, Glob) |
| 40 | Changing a file: Edit, replace_all, or Read and Write | Is the text you want to change unique in the file? | 2.5 Built-in tools (Read, Write, Edit, Bash, Grep, Glob) |
| Domain 3: Claude Code configuration and workflows |
| 41 | Where a convention or instruction should live in the CLAUDE.md hierarchy | Who needs this guidance, who owns it, and which files does it follow? | 3.1 CLAUDE.md hierarchy and scope |
| 42 | Instructions ignored or missing for some people: diagnosing which memory file is at fault | Is the guidance actually loaded, and if so, which file is it coming from? | 3.1 CLAUDE.md hierarchy and scope |
| 43 | Sharing a command with the team versus keeping it personal | Who needs the command, and how will it reach them? | 3.2 Slash commands and skills |
| 44 | Command, skill, path rule or CLAUDE.md: choosing the home by when content should load | Should this load always, when certain files are edited, or only when someone asks? | 3.2 Slash commands and skills |
| 45 | Controlling a skill: forked context, allowed tools and arguments | Is the problem what the skill can do, how much it puts into my context, or what input it needs? | 3.2 Slash commands and skills |
| 46 | Path-scoped rules versus directory CLAUDE.md for file conventions | How are the files this convention covers grouped: by pattern across folders, or inside one folder? | 3.3 Path-specific rules |
| 47 | Plan mode, Explore subagent or direct execution for a change | What is still undecided about this change, and how wide is it? | 3.4 Plan mode or direct execution |
| 48 | Examples, tests or interview: telling Claude what correct means | What do you already know about the right answer, and what kind of feedback is missing? | 3.5 Iterative refinement |
| 49 | One message or several when you have multiple problems | Does the fix for one problem depend on the outcome of another? | 3.5 Iterative refinement |
| 50 | A CI step that hangs, ends empty or behaves differently from local | Which part of the invocation is wrong: mode, permissions, output shape or configuration scope? | 3.6 Claude Code in CI/CD |
| 51 | Getting output in the shape its reader needs: schema, forced tool or format examples | Who reads the output, a program or a person, and through which interface? | 3.6 Claude Code in CI/CD |
| 52 | Where CI-specific instructions live versus repo-wide ones | Who needs these instructions: every session, one CI job, one developer, or some files? | 3.6 Claude Code in CI/CD |
| 53 | Automated review misses things: reviewer, criteria or size | Is the miss caused by who reviews, what they are told to look for, or how much they review at once? | 3.6 Claude Code in CI/CD |
| 54 | Configuring several CI jobs: real-time, batch and turn limits per job | Is this job bound by latency, by cost, or does it need a tool loop? | 3.6 Claude Code in CI/CD |
| Domain 4: Prompt engineering and structured output |
| 55 | Too many false positives: write criteria or add examples | Is the standard for a finding missing, or does it exist but get applied unevenly? | 4.1 Explicit criteria and false positives |
| 56 | Fixing one noisy review category: rewrite, pause or measure | This one category keeps producing false positives: what kind of fix does it need? | 4.1 Explicit criteria and false positives |
| 57 | Letting the model say it is unsure instead of guessing | Can the model express doubt, or is it forced to commit to an answer? | 4.1 Explicit criteria and false positives |
| 58 | Rules or worked examples when output is inconsistent | Are the rules still vague, or are they clear but the hard cases keep varying? | 4.2 Few-shot examples |
| 59 | Project conventions ignored: are the rules even loading? | Are the rules actually loaded for these files, or loaded but described poorly? | 4.2 Few-shot examples |
| 60 | Extraction across many document layouts: tailor, route or teach | Do you know the layouts in advance, and are the field descriptions already clear? | 4.2 Few-shot examples |
| 61 | Broken JSON or wrong value: what a schema does and doesn't catch | Is the structured output wrong in its syntax, its format, or its meaning? | 4.3 Structured output and JSON schemas |
| 62 | Value absent, unlisted or unreadable: which schema escape hatch | What does the document actually say about this field: nothing, something the list doesn't cover, or something ambiguous? | 4.3 Structured output and JSON schemas |
| 63 | Making the model call the right tool: tool_choice settings | Must the reply be a tool call, and if so, which tool exactly? | 4.3 Structured output and JSON schemas |
| 64 | Validation failed: retry with feedback or hand off to a person | Is the answer actually in the document for a retry to find? | 4.4 Validation, retry and feedback |
| 65 | Catching tidy-but-wrong values: where the check must live | Could the model hide this mismatch, and what outside check can catch a wrong value that looks right? | 4.4 Validation, retry and feedback |
| 66 | Measuring which review rules produce false positives | Who supplies the judgement of whether a finding was wrong: developer actions, or labelled samples? | 4.4 Validation, retry and feedback |
| 67 | Batch or real-time: who is waiting, and by when | Is anyone blocked on the result or bound by a deadline, and does each item need a tool loop? | 4.5 Batch processing |
| 68 | Working inside batch limits: tool calls, matching results, failures | What can a batch request not do, and how do you tie each result back to its input? | 4.5 Batch processing |
| 69 | Cutting cost without losing coverage | Which cost levers keep the output complete? | 4.5 Batch processing |
| 70 | Why a review misses bugs: self-review, dilution or unclear criteria | What shape are the misses, and what differs between a reviewer that finds them and one that doesn't? | 4.6 Multi-pass review |
| Domain 5: Context management and reliability |
| 71 | Keeping exact facts intact through a long conversation | What is losing the facts: the summary, the tool output, or where they sit? | 5.1 Keeping key facts in long interactions |
| 72 | Getting the middle of a long input read | Where do the missed details sit, and is the input already structured? | 5.1 Keeping key facts in long interactions |
| 73 | When the agent should hand over to a person | Which escalation trigger is actually present? | 5.2 Escalation and ambiguity |
| 74 | Fixing an escalation boundary that misfires | What is the cheapest fix that makes the escalation boundary clear? | 5.2 Escalation and ambiguity |
| 75 | Which customer record to act on | What is the source of truth for who this customer is? | 5.2 Escalation and ambiguity |
| 76 | Handling and reporting a subagent's failed or empty lookup | Did the lookup actually run, and should the subagent handle the error itself or hand it up? | 5.3 Errors across agents |
| 77 | What to do when a long session fills up or degrades | Is the session just full, or has it degraded, and what is flooding it? | 5.4 Context in large codebases |
| 78 | Carrying long work across a break, crash or handover | What carries the state forward: the conversation, a file on disk, or a fresh summary? | 5.4 Context in large codebases |
| 79 | Keeping noisy search output out of the main context | Is the noise predictable enough to exclude, or does it need filtering somewhere else? | 5.4 Context in large codebases |
| 80 | Proving a category is ready to skip human review | Which number do you actually have, and does it cover the category you plan to automate? | 5.5 Human review and confidence |
| 81 | Finding where a citation lost its source | At which stage did the claim lose the link to its source? | 5.6 Sources, provenance and uncertainty |
| 82 | Reporting figures when sources disagree | Is this a genuine conflict or a change over time, and what does the reader need? | 5.6 Sources, provenance and uncertainty |
| 83 | Speeding up human review of extracted values | Where is the reviewers' time going, and what would save it? | 5.6 Sources, provenance and uncertainty |