Pack the right inputs

For plan and ship a personal knowledge assistant in staged, testable layers, assemble only what changes the answer: one user problem, success metric, constraints, architecture sketch, milestone dependencies, evaluation set, risk register, and weekly capacity. Label each item by authority and date. A source-of-truth document outranks a memory-based note; a current error log outranks a description of last month's behavior. State conflicts instead of letting the model blend them.

Use a four-part packet: task, evidence, constraints, and output contract. Put untrusted content inside clear delimiters and say that it is data, not instruction. Include representative examples, especially one normal case and one boundary case. Omit irrelevant history; excess context can hide the one line that controls the result.

Turn this capstone goal into six two-week milestones: an assistant answers questions from my approved notes with citations. Sequence API, structured output, retrieval, citations, tools, memory, evals, injection defense, tracing, retries, deploy, and monitoring. For each milestone define a demo, test, dependency, risk, and stop rule. Do not pretend all layers fit at once.

Before sending, annotate the packet. Mark which values are verified, which are illustrative, and which are unknown. If a screenshot is involved, transcribe critical small text. If structured data is involved, include headers, units, software version, and null behavior. If creative material is involved, record ownership and permitted use. This is how context becomes operational rather than decorative.

A useful response would look like this: A dependency-aware roadmap that ships a narrow cited-answer slice early, then adds quality and operational layers behind explicit gates. That description is intentionally observable. “Looks good” is not acceptance. The operator must demo each milestone against golden tasks, inspect citations, run eval gates, test hostile inputs, review traces, exercise retries, and rehearse rollback. Keep the source material beside the draft so review means comparison, not memory.

The privacy boundary is specific: use synthetic or personally owned documents initially; separate secrets, user content, telemetry, and model prompts with least-privilege access. Create the smallest synthetic example that preserves the problem. Replace names and identifiers consistently so relationships remain testable. Redaction is not merely drawing a box: crop surrounding notifications, remove metadata where relevant, and check that hidden sheets, comments, or revision history are not included.

Poor packets lead to predictable failures: building twelve disconnected demos; adding memory before correctness; no eval baseline; hidden infrastructure work; deploying without tracing or rollback. Another common failure is silently changing the source packet mid-run. Save a version or hash of the inputs beside the output, especially when another person will reproduce the work.

Rate a packet from zero to two on six dimensions: relevance, authority, recency, completeness, privacy, and reproducibility. A score below two on authority or privacy blocks the run. A low completeness score does not invite invention; it creates a question for the owner.