Resource
AI Session Handoff — Shift Change for Working Sessions
End a working session like a shift change, not an abandonment: state captured, decisions logged, next step named — ready for the next session to pick up.
Preserve Project Decisions — Settled Stays Settled
New sessions love to reopen settled questions. Forensic fidelity carries every decision verbatim — and overriding one requires raising an explicit question first.
Overview
The most expensive failure of session boundaries is decision amnesia: the new chat cheerfully proposes offset pagination after three sessions of settling on cursors. This setup is decision preservation at maximum strength — a design-lock session in Forensic fidelity, where all five recorded decisions travel verbatim, the underlying assumption is first-class, and the handoff instructions add the forensic rule: a decision can be challenged only by raising an explicit question, never silently overridden. Settled stays settled, or the disagreement at least becomes visible.
How to use this resource
-
Capture decisions verbatim
The original wording travels — "cursor-based everywhere, offset rejected" stays exactly that strong.
-
Go forensic
Every decision and assumption travels; nothing is capped away.
-
Gate the overrides
The forensic rule converts silent reversals into explicit, answerable questions.
Why This Works
- Verbatim decisions resist the softening that paraphrase introduces
- The override gate makes disagreement visible instead of accidental
- Forensic completeness ends the which-decisions-made-the-cut lottery
Best for
- API and architecture design work
- Projects where decision drift caused rework
- Decision-heavy sessions feeding implementation sessions
Not for
- Recording decisions for human stakeholders — that's a summary, the Structured Summary Prompt's world
- Permanent team conventions that outlive the project phase — future Project Context Builder territory
Use cases
- Locking design decisions across implementation sessions
- Stopping the third relitigation of pagination
- Making decision challenges explicit instead of silent
FAQ
How do I keep a new AI chat from reopening decisions we already settled?
The HANDOFF INSTRUCTIONS carry each KEY DECISION verbatim (cursor pagination, RFC 7807 errors, ULIDs, and the rest) and add the forensic rule: if the model thinks a decision is wrong, it must "raise it as an explicit question — never override it." You paste the package as the first message of the new chat. The assistant you paste it into still decides how faithfully to follow it.
What does Forensic fidelity carry that a lighter setting wouldn't?
Forensic carries all five recorded decisions plus the underlying assumption verbatim, deduplicated but not paraphrased — nothing capped away — so you avoid the which-decisions-made-the-cut lottery. The package also tells the next session to quote it rather than reconstruct from memory. It's for feeding a design phase into implementation, not for recording decisions for human stakeholders (that's a summary).
More resources from Context Handoff Builder
Resource
Continue a Project in a New Chat — the Handoff Package
When the window fills, don't retype the project from memory. A handoff package carries the decisions, constraints, and pending work into the new session as its first message.
Resource
Carry Forward Open Tasks — the Work That Remains
Done work tells stories; open work needs carrying. The package separates completed from pending and resumes from the first open task.
Resources that pair well
Resource
Estimate Token Budget — Plan Before You Paste
Token budget planning for real workloads: how much of the window a transcript actually consumes, what is left for the answer, and how much headroom remains.
Resource
Message Too Long — the Fix That Doesn't Butcher Content
The "message too long" error has a structural fix: split at paragraph boundaries into sequenced chunks with wait rules, instead of pasting fragments and hoping.
Resource
Convert a Conversation to a Prompt — the Full Anatomy
What a converged conversation contains: a rejected voice, an accepted framing, a hard constraint, a format decision, and a final-version moment — each mapped to its prompt section.
Related tools
Tool
Context Handoff Builder
Continue work in a new chat without losing context — decisions, constraints, and tasks extracted into a state package.
Guides for this resource
Guide
How do I move context between two AI chats without repeating myself?
Starting a fresh AI chat shouldn't mean re-explaining the whole project. Here's how to turn the important parts of one conversation into a clean handoff you paste into the next.
Guide
How to Keep AI From Forgetting Project Decisions
Over a long project, AI keeps reopening decisions you already settled — re-suggesting the option you rejected, softening a constraint, renaming things. Here's how to turn those decisions into a block the model reads before it works.
Guide
Clean Up Stale Context Before It Misleads AI
You paste last month's project context and tell AI to continue — but half of it is superseded: the direction changed, a decision reversed, a deadline passed. The model treats the stale parts as current and aims the output wrong. Here's how to audit context before it misleads AI.