kill

name	kill
description	Pre-exit scan for this Lilo session — save memory-worthy lessons silently, commit and push routine changes, branch off anything risky. Use when the operator says "/kill", "kill the session", "wrap up", or signals the session is ending. NOT for killing PM tmux sessions.

kill

Operator is closing this Lilo session. Save what'd be lost. Take action — don't ask permission for routine stuff. The operator has pre-authorized commit + push as part of /kill.

Scan — fire in parallel

Run these in a single message with multiple tool calls so they execute in parallel. The point: the heavy reads (outbox JSONs, full git diff) burn subagent context, not Lilo's.

1. Agent: outbox-sweeper — sweep ../*/.lilo-outbox/*.json and report queued messages.
2. Agent: security-reviewer — scan the uncommitted diff for secrets, credential paths, and other risk markers (see below). Pass it the full risk-marker list and ask for a JSON-shaped report: {risky: [{path, reason}], skip: [paths]}.
3. Bash (batched): git status -s && git diff --stat && tmux ls 2>/dev/null || true — trivial reads, run inline.

While those run, scan this session's transcript for memory-worthy moments — corrections, validated unusual choices, hardware/external-system gotchas, surprising project facts. Cross-check against MEMORY.md. Save new ones directly; no permission needed. (Memory has to stay in the parent — subagents don't see Lilo's transcript.)

Act

Once the parallel phase returns, merge the findings and classify the uncommitted diff:

• Skip = .env, .env.local, credentials.*, *.pem, *.key, id_rsa*, *.p12 — never commit, even on a branch. List in output.
• Risky = whatever security-reviewer flagged, plus:
  • Changes to .github/workflows/, CI config, deploy scripts, shared infra
  • Large binaries (>5 MB) where intent is unclear
  • Anything that would require --no-verify to land
  • Anything you don't recognize and can't justify in one line
• Routine = the remainder. Skill / agent-registry / template edits, doc edits, .gitignore, new tools/ files, anything you authored this session and could explain in one sentence.

Then act:

• Outbox queue → relay anything the sweeper found before exit.
• Routine → stage, draft a one-line commit message, commit, push to the current branch (default main).
• Risky → stash, create branch kill/<short-slug> off the current commit, restore on the branch, commit, push the branch. Do NOT merge to main.
• Mixed → split per-file. Routine to main as one commit; risky to kill/<slug> as a separate commit. Both pushed.
• Tmux → list for awareness; PMs outlive Lilo, don't kill.

Risk markers reference (for the security-reviewer prompt)

• File contents matching (?i)(api[_-]?key|secret|token|bearer|password)\s*[:=]
• File path under .env*, credentials.*, *.pem, *.key, id_rsa*, *.p12
• .github/workflows/, CI config, deploy scripts
• Binaries >5 MB
• Anything triggering a precommit hook failure

Hard rules

• Never --amend a published commit.
• Never push with --force / --force-with-lease.
• Never bypass hooks. If a hook fails, fix the cause or branch off — do not pass --no-verify.
• Never commit .env* (except .env.example).

Output

Tight. One line per non-empty category. Two groups: Done: (actions taken) and Needs your call: (only when something was branched off or skipped). End with Safe to kill. or, if you isolated something onto a branch, Safe to kill — review <branch> when you're back.