Long Session Check
The context window is a workbench, not a warehouse. Compaction answers "it
does not fit right now"; memory answers "will we remember next time". Both are
lossy. The safe pattern is to get anything important into human-readable
files early, and to treat compaction as tidying the workbench, not as the only
safety net.
Companion to MCode Engineering Notes 03, "Keeping long sessions sharp".
Plain-text checklist: checklists/long-session.md.
This skill has two modes. Pick the one that matches the request.
Mode A - the session you are in right now
- Sort what is in context into three bins.
- Drop: things you can fetch again or no longer need (large file reads,
test logs, search results, abandoned attempts).
- Summarize: this-session-only story (where we are, next step, open bugs
and leads, files recently touched).
- Write to a file: anything that would hurt if lost or is needed in a
later session (decisions and their reasons, progress, todos, acceptance
criteria, long-lived rules).
- Write the third bin down now, in the project's progress or design doc,
or via
handoff-doc.
- Choose the next move:
- Same task, history just long: compact, and say what comes next and what
must be kept (for example "focus on the auth refactor, drop the test
debugging").
- New task: start a new session.
- Compacted several times and drifting, or repeating questions already
answered, or breaking earlier constraints: write a hand-off doc and start
fresh with it.
- Went the wrong way: rewind to before the mistake instead of compacting,
so the wrong attempt does not end up in the summary.
Mode B - reviewing an agent's design
Read the agent's code and configuration and answer each item with evidence
(file and line, config key, or observed behaviour):
- Are old tool results cleared before resorting to a full summary?
- Are tool calls and their results kept in pairs when results are cleared?
Many model APIs reject orphaned calls.
- Do key decisions and constraints get written to files, not only kept in
the conversation?
- Does manual compaction accept instructions about what to keep and what
comes next?
- Does automatic compaction trigger early enough to leave room for output
and for the summary itself? Can the session continue if compaction fails
(quota, smaller model)?
- Can the compaction request reuse the prompt cache? Does compaction run so
often that it keeps invalidating the cache?
- Does compaction change only what the model sees, leaving the user's visible
history intact? Is it kept out of user-facing budgets such as step limits?
- Is the model told that context is managed automatically, so it does not
stop half way to "continue next turn"?
- Is project memory (
AGENTS.md or similar) short and limited to stable
rules?
- Is memory scoped (session, project, user)? Are temporary progress and
todos kept out of long-term memory?
- When memory exceeds its injection budget, is truncation visible, and are
the most important rules guaranteed to load?
- Can a human open, edit and switch off the memory? Rules that must be
enforced belong in hooks, not in memory.
Output
For Mode A: the three bins, what you wrote to which file, and the chosen next
move. For Mode B: a table of the items with pass / fail / unknown, the
evidence, and the top fixes in priority order.
1---2name: long-session-check3description: Check how an agent or a working session handles long conversations, covering tool-result clearing, compaction, what gets written to files, when to start a new session, and how memory is scoped. Use when a long session starts forgetting constraints, after several compactions, or when reviewing an agent's compaction and memory design.4metadata:5 source: https://skillmd.com/skills/minimax-ai/long-session-check6---78# Long Session Check910The context window is a workbench, not a warehouse. Compaction answers "it11does not fit right now"; memory answers "will we remember next time". Both are12lossy. The safe pattern is to get anything important into human-readable13files early, and to treat compaction as tidying the workbench, not as the only14safety net.1516Companion to MCode Engineering Notes 03, "Keeping long sessions sharp".17Plain-text checklist: `checklists/long-session.md`.1819This skill has two modes. Pick the one that matches the request.2021## Mode A - the session you are in right now22231. **Sort what is in context into three bins.**24 - *Drop*: things you can fetch again or no longer need (large file reads,25 test logs, search results, abandoned attempts).26 - *Summarize*: this-session-only story (where we are, next step, open bugs27 and leads, files recently touched).28 - *Write to a file*: anything that would hurt if lost or is needed in a29 later session (decisions and their reasons, progress, todos, acceptance30 criteria, long-lived rules).312. **Write the third bin down now**, in the project's progress or design doc,32 or via `handoff-doc`.333. **Choose the next move:**34 - Same task, history just long: compact, and say what comes next and what35 must be kept (for example "focus on the auth refactor, drop the test36 debugging").37 - New task: start a new session.38 - Compacted several times and drifting, or repeating questions already39 answered, or breaking earlier constraints: write a hand-off doc and start40 fresh with it.41 - Went the wrong way: rewind to before the mistake instead of compacting,42 so the wrong attempt does not end up in the summary.4344## Mode B - reviewing an agent's design4546Read the agent's code and configuration and answer each item with evidence47(file and line, config key, or observed behaviour):48491. Are old tool results cleared before resorting to a full summary?502. Are tool calls and their results kept in pairs when results are cleared?51 Many model APIs reject orphaned calls.523. Do key decisions and constraints get written to files, not only kept in53 the conversation?544. Does manual compaction accept instructions about what to keep and what55 comes next?565. Does automatic compaction trigger early enough to leave room for output57 and for the summary itself? Can the session continue if compaction fails58 (quota, smaller model)?596. Can the compaction request reuse the prompt cache? Does compaction run so60 often that it keeps invalidating the cache?617. Does compaction change only what the model sees, leaving the user's visible62 history intact? Is it kept out of user-facing budgets such as step limits?638. Is the model told that context is managed automatically, so it does not64 stop half way to "continue next turn"?659. Is project memory (`AGENTS.md` or similar) short and limited to stable66 rules?6710. Is memory scoped (session, project, user)? Are temporary progress and68 todos kept out of long-term memory?6911. When memory exceeds its injection budget, is truncation visible, and are70 the most important rules guaranteed to load?7112. Can a human open, edit and switch off the memory? Rules that must be72 enforced belong in hooks, not in memory.7374## Output7576For Mode A: the three bins, what you wrote to which file, and the chosen next77move. For Mode B: a table of the items with pass / fail / unknown, the78evidence, and the top fixes in priority order.