[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7D6BWTCEGSCQMXTWPRNZTX2":3,"skill-same:01M7D6BWTCEGSCQMXTWPRNZTX2":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":14,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M7D6BWTCEGSCQMXTWPRNZTX2","end-session",null,"Close out a work session — capture what was learned into memory, bring GitHub issues, the board, human follow-ups and docs up to date, and clean the local workspace. Use when the user says \"end session\", \"wrap up\" or \"done for the day\", or when the handoff skill reaches its end-session step.","jwh3times\u002FPressGolf","jwh3times","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1965019?v=4","01M720KXPKWJRK0WCC0ZD480M7","Jerry Holland","01M7D681DDAHNGMFVZP9FW2PW7",0,"MIT","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\n# GENERATED — do not edit. Source: .agents\u002Fskills\u002Fend-session\u002FSKILL.md — regenerate with 'node scripts\u002Fsync-agents.mjs'.\nname: end-session\ndescription: Close out a work session — capture what was learned into memory, bring GitHub issues, the board, human follow-ups and docs up to date, and clean the local workspace. Use when the user says \"end session\", \"wrap up\" or \"done for the day\", or when the handoff skill reaches its end-session step.\nmetadata:\n  settings: workspace issue-tracker human-follow-up docs triage-labels\n---\n\n# End session\n\nClose out the session so the next one starts from an accurate picture instead of reconstructing\none. A session's durable output is more than its diff: what it **learned** (memory), what it\n**moved** (issues and the board), what a **human** must now do (follow-ups), what its **docs** now\nget wrong, and what it **left behind** (the workspace). Sweep those, in that order.\n\n**Announce at start:** \"I'm using the end-session skill to close out this session.\"\n\nRun shell snippets in bash (Git Bash on Windows). Where `gh` is unavailable, use the equivalent\nGitHub MCP operation: a missing CLI changes the interface, not the step.\n\nThis skill leaves shipping to the `ship` skill: no version bump, changelog section, source push,\nor PR. A finished branch is shipped first, then this runs; a mid-flight branch stays mid-flight,\nrecorded where it stands.\n\nThree rules hold throughout:\n\n- **Evidence, not recollection.** An empty bucket is a valid outcome; say so and move on. An\n  invented memory, comment, or entry is worse than none.\n- **List before deleting.** Every deletion — file, branch, worktree, container — is shown as a list\n  and waits for a yes.\n- **Ask before committing or discarding**, except where a setting pre-authorizes it.\n\n## 0. Settings\n\nThis skill reads `docs\u002Fagents\u002Fworkspace.md`, `docs\u002Fagents\u002Fissue-tracker.md`,\n`docs\u002Fagents\u002Fhuman-follow-up.md`, `docs\u002Fagents\u002Fdocs.md` and `docs\u002Fagents\u002Ftriage-labels.md`. Follow\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002FREADME.md` to read them and complete any that are missing; the\ndefinitions sit beside it. Resolve each `discover:` value once, here, and reuse it below.\n\n**Invoked by handoff:** when the arguments say `invoked by handoff; document at \u003Cpath>`, the\nhandoff document is already written and registered. Carry its path into the report, and treat the\nuser's work as continuing on the other machine, not in this checkout.\n\n## 1. Take stock\n\nEstablish what the session did before writing anything:\n\n```bash\ngit status --short --branch\ngit branch --show-current\ngit log --oneline origin\u002Fmain..HEAD\ngh pr list --author @me --state open --json number,title,headRefName\n```\n\nThen run the `Workspace status command`. Sort what the session produced into the buckets below;\neach routes to a later step.\n\n| Bucket | Step |\n| --- | --- |\n| Durable facts about how to work in this repository or environment | 2. Memory |\n| Work started, finished, blocked, discovered, or understood differently | 3. Issues |\n| Actions only a human can now take | 4. Human follow-ups |\n| Decisions, findings, and state changes the docs should hold | 5. Docs |\n| Files, branches, worktrees, containers left behind | 6. Workspace |\n\n## 2. Memory\n\nMemory lives outside the repository, one fact per file, in the memory directory the harness names\nin its instructions. Use that path exactly as given: a path derived from the checkout location\ndiffers per machine and per tool, and writes to a wrong one succeed silently and are never read. If\nthe harness names no memory directory, skip this step and say so in the report. `MEMORY.md` there\nis the index: one line per memory, never content.\n\nSave a candidate only when **a fresh session here would get it wrong without the note** and the\nrepository does not already record it. `AGENTS.md`, the docs, the issue tracker, the changelog,\nand Git history are durable; a memory restating them goes stale the moment they change. Memory is\nfor what is written nowhere: a trap that cost time, a preference the user stated, the state of\nhalf-decided work.\n\n- Read `MEMORY.md` first, and update a matching memory rather than adding a near-duplicate.\n- Correct or delete a memory this session proved wrong.\n- Write absolute dates, link related memories with `[[slug]]`, and keep every `MEMORY.md` pointer\n  in step with the files.\n- Keep an unpatched security weakness out of memory bodies; point at where it is tracked instead.\n\n## 3. Issues and the board\n\nIssues carry the work; the `Board` carries its state. Labels are the five in\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Ftriage-labels.md`, plus any `Extra labels` in\n`docs\u002Fagents\u002Ftriage-labels.md`.\n\nBoard writes need the `gh` token's `project` scope, and fail silently without it. Check\n`gh auth status` for `project` (not just `read:project`). If it is missing, report it and skip board\nwrites; the fix (`gh auth refresh -h github.com -s project`) needs a real terminal.\n\nStart from what is open, in the `Public repository` and again in the `Private issue repository`\nwhen there is one — always selected explicitly with `-R`:\n\n```bash\ngh issue list -R \u003Crepository> --state open --json number,title,labels \\\n  --jq '[.[] | {number, title, labels: [.labels[].name]}]'\n```\n\nFor each issue the session touched:\n\n- **Shipped** — comment the outcome and what changed about the understanding of the problem. Close\n  it once its PR has merged; while the PR is open, leave it open and name the PR.\n- **Advanced but unfinished** — comment where it stands and the next concrete step, written for a\n  reader arriving cold.\n- **Blocked** — comment the blocker, and set the `Board fields` that record it.\n- **Understanding changed** — correct the body, or comment the correction. An issue describing the\n  wrong problem costs more than a missing one.\n- **Newly discovered work** — record it as `New work` says: an issue, labelled and added to the\n  `Board`; or a `Board` draft item (`gh project item-create`), left for the work's commitment to\n  turn into an issue. Set its board fields either way. Work that never reaches the board is\n  invisible to the next session.\n\nA security finding follows the `Security findings` route, never a public issue before its fix\nships. Keep private-repository links and content out of public issues, PRs, and commits. Leave\nissues the session never touched to the `triage` skill.\n\n## 4. Human follow-ups\n\nRun the `## Procedure` in `.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fhuman-follow-up.md` for every\nrequired human action left by the session's completed work, including ones other skills\nalready started (docs-updater, `ship`): reconcile those rather than duplicating them.\n\n## 5. Docs\n\nRun the `docs-updater` skill on the session's work, in a subagent where the harness has one, told to\nfollow `.agents\u002Fskills\u002Fdocs-updater\u002FSKILL.md`; otherwise follow it yourself. A subagent cannot ask\nthe user, so complete its settings first, here, by following its step 0. Use `fix` mode for work not\nyet merged, and `check` mode for work `ship` already refreshed and merged, recording any drift it\nreports as new work, as step 3 does. Merged work whose `ship` report owed a wiki search gets one\nmore run in `fix` mode, limited to the wikis: those stale references are fixed, not filed.\nTell it this is `end-session`, and give it the session's discoveries as a list of changes beside\nthe diff: decisions and findings land in the `Doc map`\ndocuments whose **when** they match, honoring each one's marker. Run step 4's procedure for any\nhuman actions it returns. Its edits come back uncommitted, and join step 6's question about\nuncommitted work. Keep its report for step 7.\n\n- **Companion.** When the `Private companion` is set and its checkout exists, inspect it from its\n  own root: the outer repository's `git log`, `git show` and `git diff` cannot see inside it. When\n  the checkout is absent, report that companion docs could not be reconciled, and create no\n  directory in its place. Commit companion changes by the `## Procedure` in\n  `.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fworkspace.md`.\n- **Changelog debt.** A behavior change committed this session without its `Unreleased` entry is\n  debt to report; `ship` writes the changelog.\n- **One tracker.** A backlog, roadmap, or TODO file the session added to the tree duplicates the\n  issue tracker; offer to move its content into issues.\n\n## 6. Workspace\n\n**Uncommitted work first** — it is the one thing cleanup can lose for good:\n\n```bash\ngit status --short\ngit stash list\n```\n\nRun the same in the companion when it exists. For each change, ask: commit it on its branch, stash\nit with a named message, or leave it. Leaving it is a valid answer; say so in the report. A\ncompanion change is committed only by the `## Procedure` in\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fworkspace.md`.\n\n**Branches.** A branch is finished when its PR merged at or after the branch's local tip — true for\nsquash and merge commits alike, where `git branch --merged` misses every squashed branch:\n\n```bash\ngit fetch --prune origin\nfor branch in $(git for-each-ref --format='%(refname:short)' refs\u002Fheads | grep -vx main); do\n  tip=$(git rev-parse \"$branch\")\n  pr=$(gh pr list --state all --head \"$branch\" --json number,state,headRefOid \\\n    --jq 'sort_by(.number) | last | select(.) | \"\\(.number) \\(.state) \\(.headRefOid)\"')\n  read -r number state head \u003C\u003C\u003C \"$pr\"\n  verdict=unfinished\n  if [ \"$state\" = MERGED ]; then\n    git cat-file -e \"$head^{commit}\" 2>\u002Fdev\u002Fnull || git fetch -q origin \"pull\u002F$number\u002Fhead\"\n    git merge-base --is-ancestor \"$tip\" \"$head\" && verdict=finished\n  elif [ -z \"$number\" ] && git merge-base --is-ancestor \"$tip\" origin\u002Fmain; then\n    verdict=finished\n  fi\n  echo \"$branch ${number:-no-pr} ${state:-none} $verdict\"\ndone\ngit worktree list\n```\n\nDelete finished branches with `git branch -D` once the user agrees (`-d` refuses squashed ones). A\nbranch with an open PR, no PR and commits of its own, or commits after its merged head holds work:\nkeep it, and name `ship` in the report when it is ready. A remote branch still listed after the\nprune gets the same check before `git push origin --delete`. Remove a finished worktree with\n`git worktree remove \u003Cpath>`, then `git worktree prune`.\n\n**Return to `main`** last, after the tidying below, unless the tree has uncommitted tracked changes\nor the user is continuing on this branch in this checkout. Record `before=\"$(git rev-parse HEAD)\"`,\nrun `git switch main`, then `Sync with main`. If the sync refuses to fast-forward, local `main` has\ncommits of its own: report it and leave `main` as it is. A companion that has diverged from its\nupstream (commits on both sides) is reported, never merged, rebased, reset or force-pushed. For\neach dependency lockfile in `git diff --name-only \"$before\" HEAD`, install from that lockfile in its\nfolder, in the frozen form the repository's `AGENTS.md` gives (`npm ci`,\n`pnpm install --frozen-lockfile`, `dotnet tool restore`), asking first where `AGENTS.md` gates\ninstalls. Then run `Hook: after sync` from `docs\u002Fagents\u002Fworkspace.md`.\n\n**Files.** Delete `Safe-to-remove paths`, and stray scratch files the session left in the tree, by\nexplicit path after listing them. `Never-remove paths` stay unless the user explicitly asks;\nblanket deletes such as `git clean -x` would take them too, so delete only by path.\n\n**Then:**\n\n- `Dev stack` — report what is running and ask before stopping it; a running database may be\n  deliberate. Never run its **Never run** commands here.\n- `Generated-config check` — when it reports drift, regenerate, and commit both sides.\n- `Hook: end-session cleanup` from `docs\u002Fagents\u002Fworkspace.md`.\n- `Final portability check`, last. Name the exact state of each repository rather than calling the\n  checkout synchronized.\n\n## 7. Report\n\nRun this last, so the counts are the session's final state. Set `public_repository` from\n`Public repository` and `companion_repository` from `Private issue repository` (empty for `none`).\nFor each other repository the session wrote to (pushed to, or opened a PR or issue in), run it again\nwith that repository's own settings.\n\n```bash\ncount_open() {\n  if result=\"$(gh api --paginate \"$endpoint\" --jq \"$field\" 2>&1)\"; then\n    printf '%s' \"$result\" | grep -c .\n  else\n    case \"$result\" in\n      *disabled*|*\"not enabled\"*|*\"no analysis found\"*|*\"must be enabled\"*) echo disabled ;;\n      *) echo \"not readable\" ;;\n    esac\n  fi\n}\nfor column in \"public $public_repository\" \"companion $companion_repository\"; do\n  read -r role repository \u003C\u003C\u003C \"$column\"\n  [ -n \"$repository\" ] || continue\n  echo \"$role open PRs: $(gh pr list -R \"$repository\" --state open --limit 500 --json number --jq length 2>\u002Fdev\u002Fnull || echo 'not readable')\"\n  field='.[].number'\n  for kind in dependabot code-scanning secret-scanning; do\n    endpoint=\"repos\u002F$repository\u002F$kind\u002Falerts?state=open&per_page=100\"\n    echo \"$role $kind: $(count_open)\"\n  done\n  if [ \"$role\" = public ]; then\n    endpoint=\"repos\u002F$repository\u002Fsecurity-advisories?state=draft&per_page=100\"\n    field='.[].ghsa_id'\n    echo \"$role draft advisories: $(count_open)\"\n  elif ! labels=\"$(gh label list -R \"$repository\" --limit 500 --json name --jq '.[].name' 2>\u002Fdev\u002Fnull)\"; then\n    echo \"$role security issues: not readable\"\n  elif ! printf '%s\\n' \"$labels\" | grep -qx security; then\n    echo \"$role security issues: no label\"\n  else\n    gh issue list -R \"$repository\" --state open --label security --limit 500 --json labels --jq \"\n      [.[] | [.labels[].name] | any(. == \\\"wontfix\\\" or . == \\\"deferred-trigger\\\")] |\n      \\\"$role security issues: open \\(map(select(. | not)) | length), accepted risk \\(map(select(.)) | length)\\\"\" \\\n      2>\u002Fdev\u002Fnull || echo \"$role security issues: not readable\"\n  fi\ndone\n```\n\nThen print this template, every section every time, with `none` where there is nothing to say:\n\n```markdown\n## End of session: \u003Crepository>, \u003Cdate>\n\n### What this session did\n- Memory: \u003Cadded, updated or deleted, each with its reason; or why this step was skipped>\n- Issues: \u003Ccommented, closed, labelled or opened, by number>\n- Workspace: \u003Cdeleted; kept and why; uncommitted changes, unpushed branches, running containers>\n- Settings files written: \u003Clist from step 0>\n- Handoff document: \u003Cits path, when handoff invoked this skill>\n- Not done: \u003Ceverything skipped or left unfinished, and changelog debt>\n\n### Docs\n- Searched: \u003Cdocuments checked, the terms searched for, and where, from docs-updater's report>\n- Updated: \u003Cdocuments and their commit state, the companion's too, or that it was absent>\n- Drift reported, not fixed: \u003Clist>\n\n### Repository status\n| | Public | Private companion |\n|---|---|---|\n| Open PRs: this session | n | n |\n| Open PRs: others and bots | n | n |\n| Dependabot alerts | n | n |\n| Code scanning alerts | n | n |\n| Secret scanning alerts | n | n |\n| Draft advisories | n | n\u002Fa |\n| `security` issues: open | n\u002Fa | n |\n| `security` issues: accepted risk | n\u002Fa | n |\n\n### Human steps before the next session\n- \u003Cwhat, where, and why it blocks>\n\n### Human follow-ups\n- \u003Cverified issue or wiki link, or what is unpublished and why>\n```\n\n- **Status cells** hold a count, `n\u002Fa`, or what the snippet printed: `disabled`, `not readable` or\n  `no label`. Those three are never written as `0`: none of them says nothing is open. `this session`\n  counts the PRs this session opened that are still open, and `others and bots` is the rest of the\n  total. An open `security` issue also labelled `wontfix` or `deferred-trigger` is an accepted risk.\n- **Columns.** Leave out the companion column when there is no companion. Each other repository\n  gets its own pair, headed by its public name; a companion whose repository cannot be resolved has\n  `not readable` in every cell.\n- **Human steps before the next session** holds only what stops the next session: a PR to merge, an\n  expired login, work to push from this machine, a decision the user owes. Tracked work with no such\n  urgency is a follow-up.\n- **Counts and roles only.** The security rows carry no title, identifier or package name, and the\n  report names a companion by its role, never its repository. A report holding a companion's counts\n  stays out of public issues, PRs and commits.\n\nName plainly what was not done. An accurate \"these three things are still open\" is the point of the\nskill.\n","# End session\n\nClose out the session so the next one starts from an accurate picture instead of reconstructing\none. A session's durable output is more than its diff: what it **learned** (memory), what it\n**moved** (issues and the board), what a **human** must now do (follow-ups), what its **docs** now\nget wrong, and what it **left behind** (the workspace). Sweep those, in that order.\n\n**Announce at start:** \"I'm using the end-session skill to close out this session.\"\n\nRun shell snippets in bash (Git Bash on Windows). Where `gh` is unavailable, use the equivalent\nGitHub MCP operation: a missing CLI changes the interface, not the step.\n\nThis skill leaves shipping to the `ship` skill: no version bump, changelog section, source push,\nor PR. A finished branch is shipped first, then this runs; a mid-flight branch stays mid-flight,\nrecorded where it stands.\n\nThree rules hold throughout:\n\n- **Evidence, not recollection.** An empty bucket is a valid outcome; say so and move on. An\n  invented memory, comment, or entry is worse than none.\n- **List before deleting.** Every deletion — file, branch, worktree, container — is shown as a list\n  and waits for a yes.\n- **Ask before committing or discarding**, except where a setting pre-authorizes it.\n\n## 0. Settings\n\nThis skill reads `docs\u002Fagents\u002Fworkspace.md`, `docs\u002Fagents\u002Fissue-tracker.md`,\n`docs\u002Fagents\u002Fhuman-follow-up.md`, `docs\u002Fagents\u002Fdocs.md` and `docs\u002Fagents\u002Ftriage-labels.md`. Follow\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002FREADME.md` to read them and complete any that are missing; the\ndefinitions sit beside it. Resolve each `discover:` value once, here, and reuse it below.\n\n**Invoked by handoff:** when the arguments say `invoked by handoff; document at \u003Cpath>`, the\nhandoff document is already written and registered. Carry its path into the report, and treat the\nuser's work as continuing on the other machine, not in this checkout.\n\n## 1. Take stock\n\nEstablish what the session did before writing anything:\n\n```bash\ngit status --short --branch\ngit branch --show-current\ngit log --oneline origin\u002Fmain..HEAD\ngh pr list --author @me --state open --json number,title,headRefName\n```\n\nThen run the `Workspace status command`. Sort what the session produced into the buckets below;\neach routes to a later step.\n\n| Bucket | Step |\n| --- | --- |\n| Durable facts about how to work in this repository or environment | 2. Memory |\n| Work started, finished, blocked, discovered, or understood differently | 3. Issues |\n| Actions only a human can now take | 4. Human follow-ups |\n| Decisions, findings, and state changes the docs should hold | 5. Docs |\n| Files, branches, worktrees, containers left behind | 6. Workspace |\n\n## 2. Memory\n\nMemory lives outside the repository, one fact per file, in the memory directory the harness names\nin its instructions. Use that path exactly as given: a path derived from the checkout location\ndiffers per machine and per tool, and writes to a wrong one succeed silently and are never read. If\nthe harness names no memory directory, skip this step and say so in the report. `MEMORY.md` there\nis the index: one line per memory, never content.\n\nSave a candidate only when **a fresh session here would get it wrong without the note** and the\nrepository does not already record it. `AGENTS.md`, the docs, the issue tracker, the changelog,\nand Git history are durable; a memory restating them goes stale the moment they change. Memory is\nfor what is written nowhere: a trap that cost time, a preference the user stated, the state of\nhalf-decided work.\n\n- Read `MEMORY.md` first, and update a matching memory rather than adding a near-duplicate.\n- Correct or delete a memory this session proved wrong.\n- Write absolute dates, link related memories with `[[slug]]`, and keep every `MEMORY.md` pointer\n  in step with the files.\n- Keep an unpatched security weakness out of memory bodies; point at where it is tracked instead.\n\n## 3. Issues and the board\n\nIssues carry the work; the `Board` carries its state. Labels are the five in\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Ftriage-labels.md`, plus any `Extra labels` in\n`docs\u002Fagents\u002Ftriage-labels.md`.\n\nBoard writes need the `gh` token's `project` scope, and fail silently without it. Check\n`gh auth status` for `project` (not just `read:project`). If it is missing, report it and skip board\nwrites; the fix (`gh auth refresh -h github.com -s project`) needs a real terminal.\n\nStart from what is open, in the `Public repository` and again in the `Private issue repository`\nwhen there is one — always selected explicitly with `-R`:\n\n```bash\ngh issue list -R \u003Crepository> --state open --json number,title,labels \\\n  --jq '[.[] | {number, title, labels: [.labels[].name]}]'\n```\n\nFor each issue the session touched:\n\n- **Shipped** — comment the outcome and what changed about the understanding of the problem. Close\n  it once its PR has merged; while the PR is open, leave it open and name the PR.\n- **Advanced but unfinished** — comment where it stands and the next concrete step, written for a\n  reader arriving cold.\n- **Blocked** — comment the blocker, and set the `Board fields` that record it.\n- **Understanding changed** — correct the body, or comment the correction. An issue describing the\n  wrong problem costs more than a missing one.\n- **Newly discovered work** — record it as `New work` says: an issue, labelled and added to the\n  `Board`; or a `Board` draft item (`gh project item-create`), left for the work's commitment to\n  turn into an issue. Set its board fields either way. Work that never reaches the board is\n  invisible to the next session.\n\nA security finding follows the `Security findings` route, never a public issue before its fix\nships. Keep private-repository links and content out of public issues, PRs, and commits. Leave\nissues the session never touched to the `triage` skill.\n\n## 4. Human follow-ups\n\nRun the `## Procedure` in `.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fhuman-follow-up.md` for every\nrequired human action left by the session's completed work, including ones other skills\nalready started (docs-updater, `ship`): reconcile those rather than duplicating them.\n\n## 5. Docs\n\nRun the `docs-updater` skill on the session's work, in a subagent where the harness has one, told to\nfollow `.agents\u002Fskills\u002Fdocs-updater\u002FSKILL.md`; otherwise follow it yourself. A subagent cannot ask\nthe user, so complete its settings first, here, by following its step 0. Use `fix` mode for work not\nyet merged, and `check` mode for work `ship` already refreshed and merged, recording any drift it\nreports as new work, as step 3 does. Merged work whose `ship` report owed a wiki search gets one\nmore run in `fix` mode, limited to the wikis: those stale references are fixed, not filed.\nTell it this is `end-session`, and give it the session's discoveries as a list of changes beside\nthe diff: decisions and findings land in the `Doc map`\ndocuments whose **when** they match, honoring each one's marker. Run step 4's procedure for any\nhuman actions it returns. Its edits come back uncommitted, and join step 6's question about\nuncommitted work. Keep its report for step 7.\n\n- **Companion.** When the `Private companion` is set and its checkout exists, inspect it from its\n  own root: the outer repository's `git log`, `git show` and `git diff` cannot see inside it. When\n  the checkout is absent, report that companion docs could not be reconciled, and create no\n  directory in its place. Commit companion changes by the `## Procedure` in\n  `.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fworkspace.md`.\n- **Changelog debt.** A behavior change committed this session without its `Unreleased` entry is\n  debt to report; `ship` writes the changelog.\n- **One tracker.** A backlog, roadmap, or TODO file the session added to the tree duplicates the\n  issue tracker; offer to move its content into issues.\n\n## 6. Workspace\n\n**Uncommitted work first** — it is the one thing cleanup can lose for good:\n\n```bash\ngit status --short\ngit stash list\n```\n\nRun the same in the companion when it exists. For each change, ask: commit it on its branch, stash\nit with a named message, or leave it. Leaving it is a valid answer; say so in the report. A\ncompanion change is committed only by the `## Procedure` in\n`.agents\u002Fskills\u002Fend-session\u002Fsettings\u002Fworkspace.md`.\n\n**Branches.** A branch is finished when its PR merged at or after the branch's local tip — true for\nsquash and merge commits alike, where `git branch --merged` misses every squashed branch:\n\n```bash\ngit fetch --prune origin\nfor branch in $(git for-each-ref --format='%(refname:short)' refs\u002Fheads | grep -vx main); do\n  tip=$(git rev-parse \"$branch\")\n  pr=$(gh pr list --state all --head \"$branch\" --json number,state,headRefOid \\\n    --jq 'sort_by(.number) | last | select(.) | \"\\(.number) \\(.state) \\(.headRefOid)\"')\n  read -r number state head \u003C\u003C\u003C \"$pr\"\n  verdict=unfinished\n  if [ \"$state\" = MERGED ]; then\n    git cat-file -e \"$head^{commit}\" 2>\u002Fdev\u002Fnull || git fetch -q origin \"pull\u002F$number\u002Fhead\"\n    git merge-base --is-ancestor \"$tip\" \"$head\" && verdict=finished\n  elif [ -z \"$number\" ] && git merge-base --is-ancestor \"$tip\" origin\u002Fmain; then\n    verdict=finished\n  fi\n  echo \"$branch ${number:-no-pr} ${state:-none} $verdict\"\ndone\ngit worktree list\n```\n\nDelete finished branches with `git branch -D` once the user agrees (`-d` refuses squashed ones). A\nbranch with an open PR, no PR and commits of its own, or commits after its merged head holds work:\nkeep it, and name `ship` in the report when it is ready. A remote branch still listed after the\nprune gets the same check before `git push origin --delete`. Remove a finished worktree with\n`git worktree remove \u003Cpath>`, then `git worktree prune`.\n\n**Return to `main`** last, after the tidying below, unless the tree has uncommitted tracked changes\nor the user is continuing on this branch in this checkout. Record `before=\"$(git rev-parse HEAD)\"`,\nrun `git switch main`, then `Sync with main`. If the sync refuses to fast-forward, local `main` has\ncommits of its own: report it and leave `main` as it is. A companion that has diverged from its\nupstream (commits on both sides) is reported, never merged, rebased, reset or force-pushed. For\neach dependency lockfile in `git diff --name-only \"$before\" HEAD`, install from that lockfile in its\nfolder, in the frozen form the repository's `AGENTS.md` gives (`npm ci`,\n`pnpm install --frozen-lockfile`, `dotnet tool restore`), asking first where `AGENTS.md` gates\ninstalls. Then run `Hook: after sync` from `docs\u002Fagents\u002Fworkspace.md`.\n\n**Files.** Delete `Safe-to-remove paths`, and stray scratch files the session left in the tree, by\nexplicit path after listing them. `Never-remove paths` stay unless the user explicitly asks;\nblanket deletes such as `git clean -x` would take them too, so delete only by path.\n\n**Then:**\n\n- `Dev stack` — report what is running and ask before stopping it; a running database may be\n  deliberate. Never run its **Never run** commands here.\n- `Generated-config check` — when it reports drift, regenerate, and commit both sides.\n- `Hook: end-session cleanup` from `docs\u002Fagents\u002Fworkspace.md`.\n- `Final portability check`, last. Name the exact state of each repository rather than calling the\n  checkout synchronized.\n\n## 7. Report\n\nRun this last, so the counts are the session's final state. Set `public_repository` from\n`Public repository` and `companion_repository` from `Private issue repository` (empty for `none`).\nFor each other repository the session wrote to (pushed to, or opened a PR or issue in), run it again\nwith that repository's own settings.\n\n```bash\ncount_open() {\n  if result=\"$(gh api --paginate \"$endpoint\" --jq \"$field\" 2>&1)\"; then\n    printf '%s' \"$result\" | grep -c .\n  else\n    case \"$result\" in\n      *disabled*|*\"not enabled\"*|*\"no analysis found\"*|*\"must be enabled\"*) echo disabled ;;\n      *) echo \"not readable\" ;;\n    esac\n  fi\n}\nfor column in \"public $public_repository\" \"companion $companion_repository\"; do\n  read -r role repository \u003C\u003C\u003C \"$column\"\n  [ -n \"$repository\" ] || continue\n  echo \"$role open PRs: $(gh pr list -R \"$repository\" --state open --limit 500 --json number --jq length 2>\u002Fdev\u002Fnull || echo 'not readable')\"\n  field='.[].number'\n  for kind in dependabot code-scanning secret-scanning; do\n    endpoint=\"repos\u002F$repository\u002F$kind\u002Falerts?state=open&per_page=100\"\n    echo \"$role $kind: $(count_open)\"\n  done\n  if [ \"$role\" = public ]; then\n    endpoint=\"repos\u002F$repository\u002Fsecurity-advisories?state=draft&per_page=100\"\n    field='.[].ghsa_id'\n    echo \"$role draft advisories: $(count_open)\"\n  elif ! labels=\"$(gh label list -R \"$repository\" --limit 500 --json name --jq '.[].name' 2>\u002Fdev\u002Fnull)\"; then\n    echo \"$role security issues: not readable\"\n  elif ! printf '%s\\n' \"$labels\" | grep -qx security; then\n    echo \"$role security issues: no label\"\n  else\n    gh issue list -R \"$repository\" --state open --label security --limit 500 --json labels --jq \"\n      [.[] | [.labels[].name] | any(. == \\\"wontfix\\\" or . == \\\"deferred-trigger\\\")] |\n      \\\"$role security issues: open \\(map(select(. | not)) | length), accepted risk \\(map(select(.)) | length)\\\"\" \\\n      2>\u002Fdev\u002Fnull || echo \"$role security issues: not readable\"\n  fi\ndone\n```\n\nThen print this template, every section every time, with `none` where there is nothing to say:\n\n```markdown\n## End of session: \u003Crepository>, \u003Cdate>\n\n### What this session did\n- Memory: \u003Cadded, updated or deleted, each with its reason; or why this step was skipped>\n- Issues: \u003Ccommented, closed, labelled or opened, by number>\n- Workspace: \u003Cdeleted; kept and why; uncommitted changes, unpushed branches, running containers>\n- Settings files written: \u003Clist from step 0>\n- Handoff document: \u003Cits path, when handoff invoked this skill>\n- Not done: \u003Ceverything skipped or left unfinished, and changelog debt>\n\n### Docs\n- Searched: \u003Cdocuments checked, the terms searched for, and where, from docs-updater's report>\n- Updated: \u003Cdocuments and their commit state, the companion's too, or that it was absent>\n- Drift reported, not fixed: \u003Clist>\n\n### Repository status\n| | Public | Private companion |\n|---|---|---|\n| Open PRs: this session | n | n |\n| Open PRs: others and bots | n | n |\n| Dependabot alerts | n | n |\n| Code scanning alerts | n | n |\n| Secret scanning alerts | n | n |\n| Draft advisories | n | n\u002Fa |\n| `security` issues: open | n\u002Fa | n |\n| `security` issues: accepted risk | n\u002Fa | n |\n\n### Human steps before the next session\n- \u003Cwhat, where, and why it blocks>\n\n### Human follow-ups\n- \u003Cverified issue or wiki link, or what is unpublished and why>\n```\n\n- **Status cells** hold a count, `n\u002Fa`, or what the snippet printed: `disabled`, `not readable` or\n  `no label`. Those three are never written as `0`: none of them says nothing is open. `this session`\n  counts the PRs this session opened that are still open, and `others and bots` is the rest of the\n  total. An open `security` issue also labelled `wontfix` or `deferred-trigger` is an accepted risk.\n- **Columns.** Leave out the companion column when there is no companion. Each other repository\n  gets its own pair, headed by its public name; a companion whose repository cannot be resolved has\n  `not readable` in every cell.\n- **Human steps before the next session** holds only what stops the next session: a PR to merge, an\n  expired login, work to push from this machine, a decision the user owes. Tracked work with no such\n  urgency is a follow-up.\n- **Counts and roles only.** The security rows carry no title, identifier or package name, and the\n  report names a companion by its role, never its repository. A report holding a companion's counts\n  stays out of public issues, PRs and commits.\n\nName plainly what was not done. An accurate \"these three things are still open\" is the point of the\nskill.\n",{"name":5,"metadata":24,"description":7},{"settings":25},"workspace issue-tracker human-follow-up docs triage-labels","main","ac44e1040314460343f84f4bd5f1aa85b4b4b876",".claude\u002Fskills\u002Fend-session","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf","https:\u002F\u002Fgithub.com\u002Fjwh3times","2026-11-14 02:59:14",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add jwh3times\u002FPressGolf --skill end-session",{"tool":38,"command":39},"bunx","bunx skills add jwh3times\u002FPressGolf --skill end-session",{"tool":41,"command":42},"pnpm","pnpm dlx skills add jwh3times\u002FPressGolf --skill end-session","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002FSKILL.md",true,[46,52,56,60,64,68],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"settings\u002Fdocs.md","REFERENCE","参考文档",3405,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002Fdocs.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"settings\u002Fhuman-follow-up.md",6331,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002Fhuman-follow-up.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"settings\u002Fissue-tracker.md",4028,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002Fissue-tracker.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"settings\u002FREADME.md",3227,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002FREADME.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"settings\u002Ftriage-labels.md",1236,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002Ftriage-labels.md",{"path":69,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":70,"sourceHtmlUrl":71},"settings\u002Fworkspace.md",7151,"https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fend-session\u002Fsettings\u002Fworkspace.md",[73,80,87,94,101,108],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7D6DTV21VG4SEP9HSJ19J7X","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7D6DQM6J1N47BA0XN9RKFXB","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwizard\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7D6DKAWVXENK9HVDPZ2G118","what-next","Report what to work on next in this repository — an active handoff waiting to be resumed, work already in flight, ready issues ranked the way the board ranks them, what needs the user, what needs shaping first, and blocks that have quietly cleared — and recommend the one command to run. Read-only. Use when the user asks what to work on next, what is ready or what to pick up, or when a phase ends with no next piece of work chosen.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwhat-next\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7D6DFPQP3V2N3XAGPFMZPTJ","wayfinder","Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwayfinder\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":19},"01M7D6DE1118VKEQ13QZDVQ5F4","wait-what","Stop. That last message did not land: re-pitch it.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwait-what\u002FSKILL.md",[],[],{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":19},"01M7D6DBPCBGNZ7M6CNZZ5ZDKT","triage","Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.","https:\u002F\u002Fgithub.com\u002Fjwh3times\u002FPressGolf\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftriage\u002FSKILL.md",[],[]]