Manage multi-phase projects using separate sessions per phase instead of compact, with durable handoff files.
SkillScore95Maintained
Automatic
◆Coding pack
Last verified 8/3/2026
What it changes
Passed 21 of 22 test cases with the skill, and 13 of 22 without it.
Without skill59%
With skill95%
Checks it passed that the model alone missed
The response specifies reading the session handoff file in full as the first action
The handoff document includes a dedicated section for gotchas or pitfalls discovered
The handoff document includes a dedicated section for architecture or known constraints
and 7 more in the full report
8 fixed13 passed either way1 still failing0 broke
Pass-rate lift
+36 pp
Tokens
+65%
Agent turns
0%
Response time
-45%
Model gemini-3.6-flashMeasured 8/3/2026v121 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
We finished writing code for Phase 1. Before writing the handoff file and ending the session, what git action must be completed with the codebase?
What was checked, without and with the skill
Checked
Without skill
With skill
The response requires that all code work must be committed and merged per project CI rules before session closure
Missed
Passed
case-13 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md125 linessingle file
Multi-Phase Session Management
Large software projects, long-term initiatives, or phased work—where each phase is a distinct milestone with clear acceptance criteria—benefit from a session-per-phase model instead of trying to compact context within a single conversation.
Why separate sessions per phase
AI context quality degrades measurably around 60% of capacity (before auto-compact), and subsequent compacts introduce truncation and lossy summarization. A project split across multiple compact cycles accumulates error.
Better pattern:
One AI session per phase
Each phase terminates with a commit, updated handoff, and session closure
Next phase opens fresh with the handoff file and project docs as context source
The codebase, specification documents, and handoff file are the canonical context—not the previous chat history
This preserves quality because:
The next agent (or next session with the same agent) reads living docs, not truncated chat summaries
Decisions, blocked items, and state are recorded in files meant for future readers, not buried in chat
Each phase is independently validated before moving forward
Context isn't split across a "what I remember from chat" and "what the files say"
Session lifecycle per phase
Phase completion checklist
Before closing a session:
Code is committed — all work merged to the appropriate branch or merged to main (per project CI rules)
Acceptance criteria are met — the phase has a defined "done" state; validate it
Handoff is written/updated — if the project uses session-handoff.md or equivalent, it reflects current state
Next phase is clear — the handoff names the next objective, expected duration, and any blockers
CHANGELOG or progress log is updated — significant milestones are recorded for future reference
Opening a new session for the next phase
Read the handoff first — open session-handoff.md or equivalent in full before proceeding
Check git state — git status, git branch, git log --oneline -5 to understand current state
Verify acceptance criteria from prior phase — confirm the last phase actually completed and is clean