[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CYCQ712C8S8X2QFETDCZ2K":3,"skill-same:01M7CYCQ712C8S8X2QFETDCZ2K":48},{"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":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":47},"01M7CYCQ712C8S8X2QFETDCZ2K","issue-triage",null,"Triages one, several or all untriaged tracker issues: type, code area, duplicates and possible security reports, as a report for human review. Once approved, applies a type label (`bug` or `enhancement`), a `state\u002F*` label and a plain-language comment for the reporter, or moves a feature request to Discussions where the project wants that. A `decisions` run walks through issues waiting on a maintainer. Use to classify a backlog, or one issue before acting on it. Owns the plugin's state labels, its tracker comment format, its rule for handling security reports, and when to split an issue.","mattbrailsford\u002Fumbraco-devkit","mattbrailsford","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F527305?v=4","01M7CY974D4XX1P90YSXQC1T39","Matt Brailsford","01M7CY9751PSRA3EHBH03G936J",3,"MIT","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fissue-triage\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: issue-triage\ndescription: >-\n  Triages one, several or all untriaged tracker issues: type, code area, duplicates and\n  possible security reports, as a report for human review. Once approved, applies a type\n  label (`bug` or `enhancement`), a `state\u002F*` label and a plain-language comment for the\n  reporter, or moves a feature request to Discussions where the project wants that. A `decisions` run walks through\n  issues waiting on a maintainer. Use to classify a backlog, or one issue before acting on\n  it. Owns the plugin's state labels, its tracker comment format, its rule for handling\n  security reports, and when to split an issue.\nuser-invocable: true\nargument-hint: \"[issue ID | ID range | decisions | leave blank for whole tracker]\"\n---\n\n# issue-triage\n\nRead the room on a bug tracker before anyone commits to fixing anything. Classifies, checks\nfor duplicates, points at the likely code, and flags anything that smells like a security\nreport — then hands back a report. The tracker is the only record of what's been triaged,\nand nothing is written to it until the user says so.\n\n## Scope\n\n- IN: fetch, classify, find the likely code area, duplicate-check, security-flag, report,\n  draft a comment for the reporter, and apply the state labels, comments, or closes the user\n  approves.\n- OUT: any tracker write the user hasn't approved, proving a bug's cause, writing fixes,\n  planning a feature. Once a triaged issue turns into real work, hand off to `issue-resolve`.\n\n## Preflight\n\n1. **Resolve the tracker.** Read `CLAUDE.md` for an `## Issue tracker` section (type,\n   location, access method). If it's missing, run the interview below and append the section —\n   don't guess a tracker or invent an API to call.\n2. Note whether the tracker lives in this repo or elsewhere (a separate GitHub repo, a Jira\n   project, an Azure DevOps org) — the `Location` line answers that. Every fetch uses it\n   explicitly; never assume the tracker is wherever `origin` points.\n3. Resolve the access method, in order of preference: an already-connected MCP server for that\n   tracker type, else its CLI (`gh` for GitHub, `jira` CLI, `az boards`), else a documented\n   REST call. Record whichever one actually worked in `## Issue tracker` so the next run skips\n   this detection step.\n4. Resolve the label names and where feature requests go: the defaults in Type labels and\n   State labels below, or what `CLAUDE.md` records under `## Issue tracker`. If that section\n   exists but has no `Type labels`, `State labels` or `Feature requests` line, ask just that\n   interview question and add the line. If a mapping has no entry for one of the labels, ask\n   about just that label.\n5. Resolve which issues this run covers, from the argument:\n   - One ID → that issue, whatever its labels. Naming an issue means re-triage it.\n   - A range (e.g. `120-135`) → every ID in that range that still exists and is open.\n   - `decisions` → every open `state\u002Fneeds-decision` issue, handled by the Decision run\n     below instead of the normal dispatch.\n   - No argument → every open issue with no state label, plus:\n     - every `state\u002Fneeds-info` issue the reporter has replied to since the label went on,\n       or that has gone quiet for longer than the needs-info timeout;\n     - every `state\u002Fblocked` issue whose blocker has since closed.\n\n### Interview (only if `## Issue tracker` is missing from `CLAUDE.md`)\n\nAsk in one batch:\n1. **Tracker type** — GitHub Issues, Jira, Azure DevOps Boards, or other (name it).\n2. **Location** — the repo\u002Fproject\u002Forg identifier, and whether it's the same repo as the\n   source code or a separate one.\n3. **Access** — is there an MCP server already connected for it, or should this use the CLI or\n   a token from the environment?\n4. **State labels** — does the tracker already have labels or tags for triage state? If so,\n   map them onto the states in State labels below. If not, may this create the default\n   `state\u002F*` ones? Creating them is a tracker write, so it waits for a yes.\n5. **Type labels** — use GitHub's default `bug` and `enhancement` labels, or map them onto\n   the tracker's own type labels? If the tracker has neither, may this create them?\n6. **Feature requests** — do they stay on the tracker, or go somewhere else first, such as\n   GitHub Discussions (and which category) or a forum?\n\nAppend the answers to `CLAUDE.md`:\n```markdown\n## Issue tracker\n- Type: \u003Ctype>\n- Location: \u003Crepo\u002Fproject\u002Forg — note explicitly if separate from the source repo>\n- Access: \u003CMCP server name | CLI command | REST endpoint>\n- Type labels: default   \u003C!-- or a mapping, e.g. enhancement → type\u002Ffeature -->\n- State labels: default   \u003C!-- or a mapping, e.g. ready → status\u002Faccepted -->\n- Feature requests: issues   \u003C!-- or where they go, e.g. GitHub Discussions, Ideas category -->\n- Needs-info timeout: 30 days\n```\n\n## The dedup pre-pass (only when more than one issue is in scope)\n\nParallel subagents in the same batch can't see each other's output, so run one cheap,\ndeterministic check first, before spawning anything: pull just the title (and error\ntext\u002Fstack trace, if the tracker's list view shows one) for every issue in scope, and\nexact- or near-exact-match them against each other.\nThis is a literal text comparison, not a judgment call — flag pairs, don't resolve them.\nAttach the flag (\"possibly a duplicate of #N\") to whichever subagent prompt(s) it applies to;\nthe subagent still investigates and makes the final call, since a matching title or stack\ntrace is a hint, not a fact.\n\n## Dispatch\n\n**One issue in scope:** dispatch a single `issue-triager` subagent (Agent tool,\n`subagent_type: issue-triager`) and wait for its report.\n\n**More than one:** the main thread is the orchestrator. It does not fetch, classify, or judge\nany issue itself — it dispatches one `issue-triager` subagent per issue and only aggregates.\nBatch the dispatches (parallel Agent tool calls in one message); cap concurrency around 8–10\nat a time rather than firing the whole scope at once. For a large whole-tracker run, tell the\nuser the issue count and the batch plan before starting — a few hundred open issues means a\nfew hundred subagent calls, and that's worth flagging before it runs.\n\nEvery dispatch gets: the issue ID, the tracker's type\u002Flocation\u002Faccess method, the repo path,\nany dedup-pre-pass hint for that issue, why it's back if it already had a state label (a\nreply, the needs-info timeout, or a closed blocker), the \"What counts\" list from Security\nreports below, the Type labels section with this project's label names and `Feature requests`\nline, the \"Which state\" rules from State labels, Splitting an issue, and the Tracker comments\nsection.\nThe triager searches the tracker itself, open and closed, for earlier duplicates. It also\nchecks whether an open pull request is already linked to the issue (GitHub: the issue's\n\"Development\" links or a cross-reference in its timeline), opened by the reporter or another\noutside contributor, and reports it if so.\n\n## Aggregating and reporting\n\n1. Collect every subagent's report.\n2. Pull anything flagged as a possible security report into its own short list, handled as\n   Security reports below says.\n3. Build one combined table for everything else: issue, type, likely area, priority\u002Furgency,\n   duplicate-of (if any), recommended action, one-line why (written following\n   `plain-writing`). When a triager reports a linked open PR from the reporter or another\n   outside contributor, the recommended action is \"Review with `pr-review` #\u003CPR>\", not a\n   state label or fix route — the fix already exists, it hasn't been built yet.\n4. Next to each recommended action, show the tracker write it maps to (State labels below):\n   the type label, the state label, close or move, any area label fix the triager found, and\n   the comment the triager drafted. A label fix is a tracker write like any other, applied only on approval.\n5. Present the table and the security list (if non-empty) to the user. Nothing gets written\n   back to the tracker until the user says which recommendations to apply. Then apply exactly\n   those, and report what was written.\n\n## Decision run\n\nFor issues triage has already sorted, where the only thing missing is the maintainer's call.\nNo subagents: the notes already hold the facts, so the main thread reads each issue itself.\n\n1. For each issue in scope, read its latest `Notes for the decision` and everything posted\n   since: new comments, reaction counts, newly linked duplicates.\n2. Show one table: issue, need, size, risks, suggestion, and what's new since the notes. Put\n   anything new that changes the picture (a lot more demand, a workaround someone found) in\n   the suggestion, and say so. Suggest a split only as Splitting an issue below allows.\n3. Ask the user for an answer per issue:\n   - **Yes** → `state\u002Fready`.\n   - **Yes, another way** → `state\u002Fready`, with the other way as what was agreed.\n   - **No** → close as not planned.\n   - **Not yet** → no change, no comment. It comes back on the next decision run.\n4. Draft each answer's label and comment, following Tracker comments below. Apply exactly\n   the ones the user approves, as in Aggregating and reporting.\n\n## Splitting an issue\n\nThe goal of working a tracker is to clear it. Splitting an issue, or moving part of a fix\ninto a new issue, adds tracker work, so the default is one issue, fixed whole. This applies\nwherever an issue's work gets divided: a triage or decision suggestion, the start of a fix,\nand follow-ups found during one.\n\nSplit only when one of these holds, and name which:\n\n- a part needs a decision only someone else can make. If the user can make the call, ask\n  them now and keep the part in, so the work can go on. A triager can't ask, so it writes\n  the question into its notes for the decision run instead of suggesting a split;\n- a part has a different cause or need, so it would be its own issue if reported alone;\n- a part has to ship apart from the rest: a breaking change held for the next major, a\n  database change that can't go in a patch release, or another version line.\n\nThese aren't reasons: keeping a PR small, \"out of scope\", \"can do later\", a part needing\nmore design, or one part depending on another issue (that's an order to work in).\n\nA split is for tracking, never for putting work off. Each part keeps the state the whole\nhad, unless a reason above says otherwise, and is drafted ready to work: its cause and fix\nwritten down, never just \"look into this\". Show each split with its reason, and apply it\nonly on the user's yes.\n\n## Type labels\n\nEvery issue that stays open after triage gets one type label, saying what kind of issue it\nis. The state label says where it's stuck. They're separate: the type stays the same while\nthe state moves on.\n\n| Label | For |\n|---|---|\n| `bug` | Something that doesn't work the way it's meant to |\n| `enhancement` | A request for something new, or a change to how something is meant to work |\n\n- These are GitHub's default labels, so most repos already have them. Use the names in\n  `Type labels` under `## Issue tracker` when the project maps them onto its own.\n- A report of a \"bug\" in something that works as designed is an `enhancement`. Say so kindly\n  in the comment for the reporter.\n- A wrong type label already on the issue is swapped, not added to.\n- Support questions, duplicates, security reports, and issues to close get no type label.\n- Never create a type label the tracker doesn't have without the user's yes.\n\n**Feature requests elsewhere.** When `Feature requests` under `## Issue tracker` names a place\nother than the tracker, such as GitHub Discussions or a forum, an `enhancement` goes there\ninstead of getting a state label:\n\n- On GitHub, the move is the issue's \"Convert to discussion\" button. It keeps the reporter\n  as the author and carries the comments over. GitHub's API can't do it, so once the user\n  approves, post the comment, then tell the user to click the button, with the issue's link\n  and the category to pick.\n- Anywhere else, close the issue as not planned, with the comment linking to the place.\n- It stays on the tracker, sorted by \"Which state\" below, when someone on the team opened it\n  or a maintainer has already agreed to it.\n- `Feature requests: issues`, or no line at all, keeps every request on the tracker.\n\n## State labels\n\nThe tracker holds triage state, so there's no separate log to commit or keep in sync. An open\nissue with no state label is untriaged.\n\n| Label | Means | Set by |\n|---|---|---|\n| `state\u002Fneeds-investigation` | Looks real, but nobody has proven it yet | `issue-triage` |\n| `state\u002Fneeds-info` | Waiting on the reporter for detail only they can give | `issue-triage`, `investigate-bug` |\n| `state\u002Fneeds-decision` | Clear, but waiting on a maintainer to agree to it | `issue-triage` |\n| `state\u002Fblocked` | Can't be fixed here until something outside the project changes | `issue-triage`, `investigate-bug` |\n| `state\u002Fready` | Verified and ready to fix | `issue-triage`, `investigate-bug` |\n| `state\u002Fin-progress` | A fix lane has started | `issue-resolve` |\n\n**Already fixed?** Check this first. If a merged change already fixes the issue, it gets no\nstate label:\n\n- The fix is in a release that has shipped → close it as fixed, with the \"Already fixed\"\n  comment below naming that version.\n- The fix is merged but no release has it yet → leave it open, and draft the \"Already\n  fixed\" comment without the close, naming the change. With no state label it comes back on\n  the next run, which is when to check whether the release has shipped.\n\n**Which state.** Pick the first that fits:\n\n1. `state\u002Fneeds-info` only when the missing piece is something only the reporter can give:\n   their steps, versions, setup, data, or what they expected. Don't use it for anything that\n   reading the code or trying a repro could answer. That's investigation work, not the\n   reporter's. Before asking them to go check something (a log message, a setting, a response\n   body), confirm in the code that it's actually there to find. A question built on an\n   unchecked assumption (\"there's probably something in the log, can you check?\") wastes the\n   reporter's time if the code never logs it.\n2. `state\u002Fblocked` when the fix can't happen in this project until something outside it\n   changes: a CMS bug with no package workaround planned, or a dependency's release. Link\n   what it's waiting on. If a workaround in the package is planned, it isn't blocked.\n3. `state\u002Fneeds-decision` for an issue that asks for something new, or a change to how\n   something is meant to work, that's clear enough to plan but that nobody has agreed to do. Triage doesn't make\n   that call. The user can make it while approving the report: yes means `state\u002Fready`, no\n   means a close as not planned. The label is for when they don't decide yet.\n4. `state\u002Fready` only when there's nothing left to prove or decide. A bug is ready once\n   `investigate-bug` has proven its cause, or when the report and the code together leave no\n   doubt (a wrong label, a broken link, an error message pointing at the exact line). An\n   `enhancement` is ready once it's been agreed and has enough detail to plan.\n5. `state\u002Fneeds-investigation` for any other real-looking bug. When in doubt between this and\n   `state\u002Fready`, pick this. `investigate-bug` moves it on: to `state\u002Fready` once the cause\n   is proven, to `state\u002Fneeds-info` if only the reporter can fill a gap, or to a close if it\n   isn't a bug.\n\n- An issue carries one state label at a time. Setting one removes the others.\n- Duplicates and issues to close don't get a label. Close them with the tracker's own reason\n  (duplicate, not planned) where it has one, plus a comment naming the duplicate or the why.\n- Every label change and close gets a drafted comment, as Tracker comments below says.\n- A `state\u002Fneeds-info` issue with no reply from the reporter for longer than the\n  `Needs-info timeout` in `## Issue tracker` (30 days if it isn't set) gets a drafted close\n  as not planned.\n- A fixed issue leaves the label set through closing. The close is the record.\n- Use the label names recorded in `## Issue tracker` when the project maps these onto its own.\n  Never invent a label the tracker doesn't have; create one only with the user's yes.\n\n## Tracker comments\n\nThis section is the plugin's one statement of how to write to an issue. `issue-triager`,\n`investigate-bug`, and `issue-resolve` all point here.\n\nTriage is also how the reporter hears back. Every label change or close gets one comment,\ndrafted for the user to approve, with up to two parts, in this order:\n\n1. **For the reporter.** Written following `plain-writing`, for someone who\n   uses the package but has never seen its source. Leave it out when someone on the team\n   opened the issue. Cut it down to what's new when a maintainer has already told the\n   reporter the same thing in a comment; never post a second version of their reply. No file paths, class names, or stack\n   frames unless the reporter used them first. Say what was understood, what happens next,\n   and what's needed from them, if anything. Thank them where it fits. Never promise a fix\n   date or a release.\n2. **Notes.** The technical record for whoever picks the issue up next; without it, the\n   findings are lost when the chat ends. Each outcome below gives the notes' title and\n   fields. Fold the notes away where the tracker can, so the reporter sees one line they can\n   skip: on GitHub, a `\u003Cdetails>` block with `\u003Csummary>\u003Ctitle>\u003C\u002Fsummary>` and a blank line\n   after it, so the Markdown inside still renders. Where the tracker has no collapsible\n   block, use the title as a heading.\n\nRules for the notes:\n\n- Every field comes from something actually checked. Write \"none found\" or \"not checked\"\n  rather than leave a field out or guess at it.\n- Add only what's new since the issue's last notes. Link back to them rather than repeat\n  them.\n- They're public even when folded, so leave out roadmap, commercial, or customer detail.\n\n### By outcome\n\n**`state\u002Fneeds-info`**\n- Reporter: exactly what's missing and why it's needed, asked as questions they can answer\n  without the code (steps, versions, a screenshot, a log message and where to find it),\n  checked as \"Which state\" says.\n- Notes, `Notes for the next look`:\n  - **Known so far:** the symptom and setup as reported.\n  - **Checked:** what triage or investigation already looked at, and what it ruled out.\n  - **Missing:** each question asked, and what its answer would rule in or out.\n  - **Likely area:** from a search of the code, or \"none found\".\n\n**`state\u002Fneeds-investigation`**\n- Reporter: it's been read and will be looked into, but isn't confirmed as a bug yet.\n- Notes, `Notes for the investigation`:\n  - **Symptom:** expected versus actual, quoting any error word for word.\n  - **Where it runs:** CMS version, package version, database, browser, as reported.\n  - **Likely area:** from a search of the code, or \"none found\".\n  - **Related:** similar issues, open or closed, and how they differ.\n  - **Start with:** the first thing to check, and why.\n\n**`state\u002Fneeds-decision`**\n- Reporter: the request is understood and waiting on a maintainer's decision. Don't hint at\n  the answer; the suggestion goes in the notes.\n- Notes, `Notes for the decision`:\n  - **Need:** what the reporter is trying to do, which may differ from the feature they\n    asked for.\n  - **Demand:** linked duplicates and related issues, and their reaction counts.\n  - **Today:** any workaround, setting, or extension point that already gets close.\n  - **Already covered:** whether the CMS or another package does this.\n  - **Size:** small, medium, or large, with the code area it would touch.\n  - **Risks:** a breaking change, new public API to maintain, work on each supported CMS\n    major, a performance cost.\n  - **Other ways:** a smaller or different way to meet the need, such as an extension point.\n  - **Questions first:** anything to settle before it could be planned.\n  - **Suggestion:** do it, do another way, or don't, with the one or two reasons that weigh\n    most. Label it as a suggestion; the maintainer decides. Word a \"don't\" kindly, since the\n    reporter can read it.\n\n**`state\u002Fblocked`**\n- Reporter: what it's waiting on, with a link, and that it moves once that's fixed.\n- Notes, `Notes on the blocker`:\n  - **Waiting on:** the upstream issue or release, linked.\n  - **Why here can't fix it:** the cause, and why it sits outside the project.\n  - **Workarounds weighed:** each one, and why it wasn't planned.\n  - **Unblocked when:** the version or change to watch for, and how to check it landed.\n  - **Then:** what's left to do here once it has.\n\n**`state\u002Fready`**\n- Reporter: it's confirmed, what was found in plain terms, and that it's queued for a fix.\n  For a feature request, that it's agreed. If it was agreed another way, say plainly what\n  will be built instead and why.\n- Notes, `Notes for the fix`. For a bug:\n  - **Repro:** exact steps, or the test that shows it.\n  - **Cause:** the cause, at file and line, and the check that proved it.\n  - **Same cause elsewhere:** each place, or \"none found\".\n  - **Suggested fix:** one or two lines, and the test that should guard it.\n  - **Size:** a quick fix or the full pipeline, and why.\n- For an agreed feature request:\n  - **Agreed:** what the maintainer said yes to, and anything they ruled out.\n  - **Scope:** what's in and out, and the code area it touches.\n  - **Questions left:** anything still to settle while planning.\n\n**`state\u002Fin-progress`** (drafted by `issue-resolve`)\n- Reporter: work on a fix has started.\n- Notes, `Notes on the work`:\n  - **Route:** a quick fix or the pipeline, and why.\n  - **Where:** the branch, and the plan folder if there is one.\n  - **New findings:** anything learned since the ready notes, such as `investigate-bug`'s\n    report if it ran just now.\n\n**Fix landed** (drafted by `issue-resolve`)\n- Reporter: what changed, in their terms.\n- Notes, `Notes on the fix`:\n  - **Change:** the commit or PR, linked, and the area it touched.\n  - **Verified by:** the repro run through the real entry point, and its result, or \"not\n    verified end to end\" and why.\n  - **Guarded by:** the test that now covers it.\n  - **Left out:** anything deliberately not fixed, and any follow-up issues.\n\n**Already fixed**\n- Reporter: it's fixed, and the version to upgrade to. If no release has it yet, say it's\n  fixed and will be in an upcoming release, without promising a date.\n- Notes, `Notes on the fix`:\n  - **Change:** the commit or PR that fixed it, linked.\n  - **Released in:** the version, or \"not released yet\".\n\n**Duplicate**\n- Reporter: a link to the earlier issue to follow instead.\n- No notes. Add anything new this report showed to the earlier issue instead.\n\n**Close**\n- Reporter: why, kindly. For a support question, answer it or point to the docs or forum.\n  For a feature request turned down, give the reason and any workaround that exists today.\n- Notes, `Notes on the close`, only when it was looked into:\n  - **Checked:** what was tried, such as the setup it failed to reproduce on.\n  - **Reopen if:** what new detail would make it worth another look.\n\n**Moved out of the tracker** (Feature requests elsewhere, under Type labels)\n- Reporter: thank them, and say feature ideas are talked through in that place first, where\n  others can add their own use case and vote. Say where to find it. Give any workaround that\n  exists today. Don't hint at whether it will be built.\n- No notes. The idea starts fresh where it lands.\n\n**Needs-info timeout close**\n- Reporter: there was no reply, so it's closed for now, and replying with the detail will\n  reopen it.\n- No notes. The needs-info notes already say what's missing.\n\nA security report never gets a comment (Security reports below).\n\n## Security reports\n\nThis section is the plugin's one statement of this rule. `issue-triager`, `issue-resolve`,\nand `investigate-bug` all point here.\n\n**What counts.** An issue or symptom that reads like a real vulnerability: auth bypass, an\naction allowed without permission, data shown to someone who shouldn't see it, injection or\ninjected script running, secret or credential exposure, remote code execution, a CVE\nreference, or similar.\n\n**What to do.**\n\n- Stop. Don't triage, investigate, or fix it as normal work.\n- Tell the user, naming the issue but not detailing it, and point them at the\n  responsible-disclosure process.\n- Never put it in a public commit, PR, comment, or issue, and never draft a public response\n  to it. A public reference to an unpatched vulnerability is a leak, not a fix.\n- Never quote exploit detail into a report, log, or plan file.\n- Never label it, in this tracker or any other public place, and never record it in a file.\n  The chat is the only place it appears. With no label, it stays untriaged and gets flagged\n  again on the next run, which is the reminder that it's still open.\n\n## Rules\n\n- Read-only against the codebase. The only file this skill writes is the `## Issue tracker`\n  section of `CLAUDE.md`, when it or one of its lines is missing. The tracker is written to only as the next rule\n  allows.\n- Never post a comment, apply a label, or close an issue without the user picking that action\n  explicitly — issue by issue, or as a stated \"apply all.\" A draft is the default output, not\n  an action.\n- If the tracker is unreachable (auth, network, wrong location), stop and say so. Don't triage\n  a stale cached copy or guess at an issue's content.\n\n## Hand off\n\nA triaged issue that turns into real work goes to `issue-resolve` next, with its ID. That\npicks `quick-fix` or the pipeline, and drafts the tracker reply once the fix lands. A\n`state\u002Fneeds-investigation` issue can go to `investigate-bug` first instead, to prove it\nbefore anyone commits to a fix. An issue with a linked open PR goes to `pr-review` with that\nPR number instead — name the skill explicitly, never a generic \"review the PR(s)\", since a\nbusy install can drop skill descriptions from the listing and a bare mention won't route. To\nwork the whole tracker through to ready, `issue-backlog` runs triage, decisions, and\ninvestigations in order. This skill stops at \"here's what I'd do,\" it doesn't start building.\n","# issue-triage\n\nRead the room on a bug tracker before anyone commits to fixing anything. Classifies, checks\nfor duplicates, points at the likely code, and flags anything that smells like a security\nreport — then hands back a report. The tracker is the only record of what's been triaged,\nand nothing is written to it until the user says so.\n\n## Scope\n\n- IN: fetch, classify, find the likely code area, duplicate-check, security-flag, report,\n  draft a comment for the reporter, and apply the state labels, comments, or closes the user\n  approves.\n- OUT: any tracker write the user hasn't approved, proving a bug's cause, writing fixes,\n  planning a feature. Once a triaged issue turns into real work, hand off to `issue-resolve`.\n\n## Preflight\n\n1. **Resolve the tracker.** Read `CLAUDE.md` for an `## Issue tracker` section (type,\n   location, access method). If it's missing, run the interview below and append the section —\n   don't guess a tracker or invent an API to call.\n2. Note whether the tracker lives in this repo or elsewhere (a separate GitHub repo, a Jira\n   project, an Azure DevOps org) — the `Location` line answers that. Every fetch uses it\n   explicitly; never assume the tracker is wherever `origin` points.\n3. Resolve the access method, in order of preference: an already-connected MCP server for that\n   tracker type, else its CLI (`gh` for GitHub, `jira` CLI, `az boards`), else a documented\n   REST call. Record whichever one actually worked in `## Issue tracker` so the next run skips\n   this detection step.\n4. Resolve the label names and where feature requests go: the defaults in Type labels and\n   State labels below, or what `CLAUDE.md` records under `## Issue tracker`. If that section\n   exists but has no `Type labels`, `State labels` or `Feature requests` line, ask just that\n   interview question and add the line. If a mapping has no entry for one of the labels, ask\n   about just that label.\n5. Resolve which issues this run covers, from the argument:\n   - One ID → that issue, whatever its labels. Naming an issue means re-triage it.\n   - A range (e.g. `120-135`) → every ID in that range that still exists and is open.\n   - `decisions` → every open `state\u002Fneeds-decision` issue, handled by the Decision run\n     below instead of the normal dispatch.\n   - No argument → every open issue with no state label, plus:\n     - every `state\u002Fneeds-info` issue the reporter has replied to since the label went on,\n       or that has gone quiet for longer than the needs-info timeout;\n     - every `state\u002Fblocked` issue whose blocker has since closed.\n\n### Interview (only if `## Issue tracker` is missing from `CLAUDE.md`)\n\nAsk in one batch:\n1. **Tracker type** — GitHub Issues, Jira, Azure DevOps Boards, or other (name it).\n2. **Location** — the repo\u002Fproject\u002Forg identifier, and whether it's the same repo as the\n   source code or a separate one.\n3. **Access** — is there an MCP server already connected for it, or should this use the CLI or\n   a token from the environment?\n4. **State labels** — does the tracker already have labels or tags for triage state? If so,\n   map them onto the states in State labels below. If not, may this create the default\n   `state\u002F*` ones? Creating them is a tracker write, so it waits for a yes.\n5. **Type labels** — use GitHub's default `bug` and `enhancement` labels, or map them onto\n   the tracker's own type labels? If the tracker has neither, may this create them?\n6. **Feature requests** — do they stay on the tracker, or go somewhere else first, such as\n   GitHub Discussions (and which category) or a forum?\n\nAppend the answers to `CLAUDE.md`:\n```markdown\n## Issue tracker\n- Type: \u003Ctype>\n- Location: \u003Crepo\u002Fproject\u002Forg — note explicitly if separate from the source repo>\n- Access: \u003CMCP server name | CLI command | REST endpoint>\n- Type labels: default   \u003C!-- or a mapping, e.g. enhancement → type\u002Ffeature -->\n- State labels: default   \u003C!-- or a mapping, e.g. ready → status\u002Faccepted -->\n- Feature requests: issues   \u003C!-- or where they go, e.g. GitHub Discussions, Ideas category -->\n- Needs-info timeout: 30 days\n```\n\n## The dedup pre-pass (only when more than one issue is in scope)\n\nParallel subagents in the same batch can't see each other's output, so run one cheap,\ndeterministic check first, before spawning anything: pull just the title (and error\ntext\u002Fstack trace, if the tracker's list view shows one) for every issue in scope, and\nexact- or near-exact-match them against each other.\nThis is a literal text comparison, not a judgment call — flag pairs, don't resolve them.\nAttach the flag (\"possibly a duplicate of #N\") to whichever subagent prompt(s) it applies to;\nthe subagent still investigates and makes the final call, since a matching title or stack\ntrace is a hint, not a fact.\n\n## Dispatch\n\n**One issue in scope:** dispatch a single `issue-triager` subagent (Agent tool,\n`subagent_type: issue-triager`) and wait for its report.\n\n**More than one:** the main thread is the orchestrator. It does not fetch, classify, or judge\nany issue itself — it dispatches one `issue-triager` subagent per issue and only aggregates.\nBatch the dispatches (parallel Agent tool calls in one message); cap concurrency around 8–10\nat a time rather than firing the whole scope at once. For a large whole-tracker run, tell the\nuser the issue count and the batch plan before starting — a few hundred open issues means a\nfew hundred subagent calls, and that's worth flagging before it runs.\n\nEvery dispatch gets: the issue ID, the tracker's type\u002Flocation\u002Faccess method, the repo path,\nany dedup-pre-pass hint for that issue, why it's back if it already had a state label (a\nreply, the needs-info timeout, or a closed blocker), the \"What counts\" list from Security\nreports below, the Type labels section with this project's label names and `Feature requests`\nline, the \"Which state\" rules from State labels, Splitting an issue, and the Tracker comments\nsection.\nThe triager searches the tracker itself, open and closed, for earlier duplicates. It also\nchecks whether an open pull request is already linked to the issue (GitHub: the issue's\n\"Development\" links or a cross-reference in its timeline), opened by the reporter or another\noutside contributor, and reports it if so.\n\n## Aggregating and reporting\n\n1. Collect every subagent's report.\n2. Pull anything flagged as a possible security report into its own short list, handled as\n   Security reports below says.\n3. Build one combined table for everything else: issue, type, likely area, priority\u002Furgency,\n   duplicate-of (if any), recommended action, one-line why (written following\n   `plain-writing`). When a triager reports a linked open PR from the reporter or another\n   outside contributor, the recommended action is \"Review with `pr-review` #\u003CPR>\", not a\n   state label or fix route — the fix already exists, it hasn't been built yet.\n4. Next to each recommended action, show the tracker write it maps to (State labels below):\n   the type label, the state label, close or move, any area label fix the triager found, and\n   the comment the triager drafted. A label fix is a tracker write like any other, applied only on approval.\n5. Present the table and the security list (if non-empty) to the user. Nothing gets written\n   back to the tracker until the user says which recommendations to apply. Then apply exactly\n   those, and report what was written.\n\n## Decision run\n\nFor issues triage has already sorted, where the only thing missing is the maintainer's call.\nNo subagents: the notes already hold the facts, so the main thread reads each issue itself.\n\n1. For each issue in scope, read its latest `Notes for the decision` and everything posted\n   since: new comments, reaction counts, newly linked duplicates.\n2. Show one table: issue, need, size, risks, suggestion, and what's new since the notes. Put\n   anything new that changes the picture (a lot more demand, a workaround someone found) in\n   the suggestion, and say so. Suggest a split only as Splitting an issue below allows.\n3. Ask the user for an answer per issue:\n   - **Yes** → `state\u002Fready`.\n   - **Yes, another way** → `state\u002Fready`, with the other way as what was agreed.\n   - **No** → close as not planned.\n   - **Not yet** → no change, no comment. It comes back on the next decision run.\n4. Draft each answer's label and comment, following Tracker comments below. Apply exactly\n   the ones the user approves, as in Aggregating and reporting.\n\n## Splitting an issue\n\nThe goal of working a tracker is to clear it. Splitting an issue, or moving part of a fix\ninto a new issue, adds tracker work, so the default is one issue, fixed whole. This applies\nwherever an issue's work gets divided: a triage or decision suggestion, the start of a fix,\nand follow-ups found during one.\n\nSplit only when one of these holds, and name which:\n\n- a part needs a decision only someone else can make. If the user can make the call, ask\n  them now and keep the part in, so the work can go on. A triager can't ask, so it writes\n  the question into its notes for the decision run instead of suggesting a split;\n- a part has a different cause or need, so it would be its own issue if reported alone;\n- a part has to ship apart from the rest: a breaking change held for the next major, a\n  database change that can't go in a patch release, or another version line.\n\nThese aren't reasons: keeping a PR small, \"out of scope\", \"can do later\", a part needing\nmore design, or one part depending on another issue (that's an order to work in).\n\nA split is for tracking, never for putting work off. Each part keeps the state the whole\nhad, unless a reason above says otherwise, and is drafted ready to work: its cause and fix\nwritten down, never just \"look into this\". Show each split with its reason, and apply it\nonly on the user's yes.\n\n## Type labels\n\nEvery issue that stays open after triage gets one type label, saying what kind of issue it\nis. The state label says where it's stuck. They're separate: the type stays the same while\nthe state moves on.\n\n| Label | For |\n|---|---|\n| `bug` | Something that doesn't work the way it's meant to |\n| `enhancement` | A request for something new, or a change to how something is meant to work |\n\n- These are GitHub's default labels, so most repos already have them. Use the names in\n  `Type labels` under `## Issue tracker` when the project maps them onto its own.\n- A report of a \"bug\" in something that works as designed is an `enhancement`. Say so kindly\n  in the comment for the reporter.\n- A wrong type label already on the issue is swapped, not added to.\n- Support questions, duplicates, security reports, and issues to close get no type label.\n- Never create a type label the tracker doesn't have without the user's yes.\n\n**Feature requests elsewhere.** When `Feature requests` under `## Issue tracker` names a place\nother than the tracker, such as GitHub Discussions or a forum, an `enhancement` goes there\ninstead of getting a state label:\n\n- On GitHub, the move is the issue's \"Convert to discussion\" button. It keeps the reporter\n  as the author and carries the comments over. GitHub's API can't do it, so once the user\n  approves, post the comment, then tell the user to click the button, with the issue's link\n  and the category to pick.\n- Anywhere else, close the issue as not planned, with the comment linking to the place.\n- It stays on the tracker, sorted by \"Which state\" below, when someone on the team opened it\n  or a maintainer has already agreed to it.\n- `Feature requests: issues`, or no line at all, keeps every request on the tracker.\n\n## State labels\n\nThe tracker holds triage state, so there's no separate log to commit or keep in sync. An open\nissue with no state label is untriaged.\n\n| Label | Means | Set by |\n|---|---|---|\n| `state\u002Fneeds-investigation` | Looks real, but nobody has proven it yet | `issue-triage` |\n| `state\u002Fneeds-info` | Waiting on the reporter for detail only they can give | `issue-triage`, `investigate-bug` |\n| `state\u002Fneeds-decision` | Clear, but waiting on a maintainer to agree to it | `issue-triage` |\n| `state\u002Fblocked` | Can't be fixed here until something outside the project changes | `issue-triage`, `investigate-bug` |\n| `state\u002Fready` | Verified and ready to fix | `issue-triage`, `investigate-bug` |\n| `state\u002Fin-progress` | A fix lane has started | `issue-resolve` |\n\n**Already fixed?** Check this first. If a merged change already fixes the issue, it gets no\nstate label:\n\n- The fix is in a release that has shipped → close it as fixed, with the \"Already fixed\"\n  comment below naming that version.\n- The fix is merged but no release has it yet → leave it open, and draft the \"Already\n  fixed\" comment without the close, naming the change. With no state label it comes back on\n  the next run, which is when to check whether the release has shipped.\n\n**Which state.** Pick the first that fits:\n\n1. `state\u002Fneeds-info` only when the missing piece is something only the reporter can give:\n   their steps, versions, setup, data, or what they expected. Don't use it for anything that\n   reading the code or trying a repro could answer. That's investigation work, not the\n   reporter's. Before asking them to go check something (a log message, a setting, a response\n   body), confirm in the code that it's actually there to find. A question built on an\n   unchecked assumption (\"there's probably something in the log, can you check?\") wastes the\n   reporter's time if the code never logs it.\n2. `state\u002Fblocked` when the fix can't happen in this project until something outside it\n   changes: a CMS bug with no package workaround planned, or a dependency's release. Link\n   what it's waiting on. If a workaround in the package is planned, it isn't blocked.\n3. `state\u002Fneeds-decision` for an issue that asks for something new, or a change to how\n   something is meant to work, that's clear enough to plan but that nobody has agreed to do. Triage doesn't make\n   that call. The user can make it while approving the report: yes means `state\u002Fready`, no\n   means a close as not planned. The label is for when they don't decide yet.\n4. `state\u002Fready` only when there's nothing left to prove or decide. A bug is ready once\n   `investigate-bug` has proven its cause, or when the report and the code together leave no\n   doubt (a wrong label, a broken link, an error message pointing at the exact line). An\n   `enhancement` is ready once it's been agreed and has enough detail to plan.\n5. `state\u002Fneeds-investigation` for any other real-looking bug. When in doubt between this and\n   `state\u002Fready`, pick this. `investigate-bug` moves it on: to `state\u002Fready` once the cause\n   is proven, to `state\u002Fneeds-info` if only the reporter can fill a gap, or to a close if it\n   isn't a bug.\n\n- An issue carries one state label at a time. Setting one removes the others.\n- Duplicates and issues to close don't get a label. Close them with the tracker's own reason\n  (duplicate, not planned) where it has one, plus a comment naming the duplicate or the why.\n- Every label change and close gets a drafted comment, as Tracker comments below says.\n- A `state\u002Fneeds-info` issue with no reply from the reporter for longer than the\n  `Needs-info timeout` in `## Issue tracker` (30 days if it isn't set) gets a drafted close\n  as not planned.\n- A fixed issue leaves the label set through closing. The close is the record.\n- Use the label names recorded in `## Issue tracker` when the project maps these onto its own.\n  Never invent a label the tracker doesn't have; create one only with the user's yes.\n\n## Tracker comments\n\nThis section is the plugin's one statement of how to write to an issue. `issue-triager`,\n`investigate-bug`, and `issue-resolve` all point here.\n\nTriage is also how the reporter hears back. Every label change or close gets one comment,\ndrafted for the user to approve, with up to two parts, in this order:\n\n1. **For the reporter.** Written following `plain-writing`, for someone who\n   uses the package but has never seen its source. Leave it out when someone on the team\n   opened the issue. Cut it down to what's new when a maintainer has already told the\n   reporter the same thing in a comment; never post a second version of their reply. No file paths, class names, or stack\n   frames unless the reporter used them first. Say what was understood, what happens next,\n   and what's needed from them, if anything. Thank them where it fits. Never promise a fix\n   date or a release.\n2. **Notes.** The technical record for whoever picks the issue up next; without it, the\n   findings are lost when the chat ends. Each outcome below gives the notes' title and\n   fields. Fold the notes away where the tracker can, so the reporter sees one line they can\n   skip: on GitHub, a `\u003Cdetails>` block with `\u003Csummary>\u003Ctitle>\u003C\u002Fsummary>` and a blank line\n   after it, so the Markdown inside still renders. Where the tracker has no collapsible\n   block, use the title as a heading.\n\nRules for the notes:\n\n- Every field comes from something actually checked. Write \"none found\" or \"not checked\"\n  rather than leave a field out or guess at it.\n- Add only what's new since the issue's last notes. Link back to them rather than repeat\n  them.\n- They're public even when folded, so leave out roadmap, commercial, or customer detail.\n\n### By outcome\n\n**`state\u002Fneeds-info`**\n- Reporter: exactly what's missing and why it's needed, asked as questions they can answer\n  without the code (steps, versions, a screenshot, a log message and where to find it),\n  checked as \"Which state\" says.\n- Notes, `Notes for the next look`:\n  - **Known so far:** the symptom and setup as reported.\n  - **Checked:** what triage or investigation already looked at, and what it ruled out.\n  - **Missing:** each question asked, and what its answer would rule in or out.\n  - **Likely area:** from a search of the code, or \"none found\".\n\n**`state\u002Fneeds-investigation`**\n- Reporter: it's been read and will be looked into, but isn't confirmed as a bug yet.\n- Notes, `Notes for the investigation`:\n  - **Symptom:** expected versus actual, quoting any error word for word.\n  - **Where it runs:** CMS version, package version, database, browser, as reported.\n  - **Likely area:** from a search of the code, or \"none found\".\n  - **Related:** similar issues, open or closed, and how they differ.\n  - **Start with:** the first thing to check, and why.\n\n**`state\u002Fneeds-decision`**\n- Reporter: the request is understood and waiting on a maintainer's decision. Don't hint at\n  the answer; the suggestion goes in the notes.\n- Notes, `Notes for the decision`:\n  - **Need:** what the reporter is trying to do, which may differ from the feature they\n    asked for.\n  - **Demand:** linked duplicates and related issues, and their reaction counts.\n  - **Today:** any workaround, setting, or extension point that already gets close.\n  - **Already covered:** whether the CMS or another package does this.\n  - **Size:** small, medium, or large, with the code area it would touch.\n  - **Risks:** a breaking change, new public API to maintain, work on each supported CMS\n    major, a performance cost.\n  - **Other ways:** a smaller or different way to meet the need, such as an extension point.\n  - **Questions first:** anything to settle before it could be planned.\n  - **Suggestion:** do it, do another way, or don't, with the one or two reasons that weigh\n    most. Label it as a suggestion; the maintainer decides. Word a \"don't\" kindly, since the\n    reporter can read it.\n\n**`state\u002Fblocked`**\n- Reporter: what it's waiting on, with a link, and that it moves once that's fixed.\n- Notes, `Notes on the blocker`:\n  - **Waiting on:** the upstream issue or release, linked.\n  - **Why here can't fix it:** the cause, and why it sits outside the project.\n  - **Workarounds weighed:** each one, and why it wasn't planned.\n  - **Unblocked when:** the version or change to watch for, and how to check it landed.\n  - **Then:** what's left to do here once it has.\n\n**`state\u002Fready`**\n- Reporter: it's confirmed, what was found in plain terms, and that it's queued for a fix.\n  For a feature request, that it's agreed. If it was agreed another way, say plainly what\n  will be built instead and why.\n- Notes, `Notes for the fix`. For a bug:\n  - **Repro:** exact steps, or the test that shows it.\n  - **Cause:** the cause, at file and line, and the check that proved it.\n  - **Same cause elsewhere:** each place, or \"none found\".\n  - **Suggested fix:** one or two lines, and the test that should guard it.\n  - **Size:** a quick fix or the full pipeline, and why.\n- For an agreed feature request:\n  - **Agreed:** what the maintainer said yes to, and anything they ruled out.\n  - **Scope:** what's in and out, and the code area it touches.\n  - **Questions left:** anything still to settle while planning.\n\n**`state\u002Fin-progress`** (drafted by `issue-resolve`)\n- Reporter: work on a fix has started.\n- Notes, `Notes on the work`:\n  - **Route:** a quick fix or the pipeline, and why.\n  - **Where:** the branch, and the plan folder if there is one.\n  - **New findings:** anything learned since the ready notes, such as `investigate-bug`'s\n    report if it ran just now.\n\n**Fix landed** (drafted by `issue-resolve`)\n- Reporter: what changed, in their terms.\n- Notes, `Notes on the fix`:\n  - **Change:** the commit or PR, linked, and the area it touched.\n  - **Verified by:** the repro run through the real entry point, and its result, or \"not\n    verified end to end\" and why.\n  - **Guarded by:** the test that now covers it.\n  - **Left out:** anything deliberately not fixed, and any follow-up issues.\n\n**Already fixed**\n- Reporter: it's fixed, and the version to upgrade to. If no release has it yet, say it's\n  fixed and will be in an upcoming release, without promising a date.\n- Notes, `Notes on the fix`:\n  - **Change:** the commit or PR that fixed it, linked.\n  - **Released in:** the version, or \"not released yet\".\n\n**Duplicate**\n- Reporter: a link to the earlier issue to follow instead.\n- No notes. Add anything new this report showed to the earlier issue instead.\n\n**Close**\n- Reporter: why, kindly. For a support question, answer it or point to the docs or forum.\n  For a feature request turned down, give the reason and any workaround that exists today.\n- Notes, `Notes on the close`, only when it was looked into:\n  - **Checked:** what was tried, such as the setup it failed to reproduce on.\n  - **Reopen if:** what new detail would make it worth another look.\n\n**Moved out of the tracker** (Feature requests elsewhere, under Type labels)\n- Reporter: thank them, and say feature ideas are talked through in that place first, where\n  others can add their own use case and vote. Say where to find it. Give any workaround that\n  exists today. Don't hint at whether it will be built.\n- No notes. The idea starts fresh where it lands.\n\n**Needs-info timeout close**\n- Reporter: there was no reply, so it's closed for now, and replying with the detail will\n  reopen it.\n- No notes. The needs-info notes already say what's missing.\n\nA security report never gets a comment (Security reports below).\n\n## Security reports\n\nThis section is the plugin's one statement of this rule. `issue-triager`, `issue-resolve`,\nand `investigate-bug` all point here.\n\n**What counts.** An issue or symptom that reads like a real vulnerability: auth bypass, an\naction allowed without permission, data shown to someone who shouldn't see it, injection or\ninjected script running, secret or credential exposure, remote code execution, a CVE\nreference, or similar.\n\n**What to do.**\n\n- Stop. Don't triage, investigate, or fix it as normal work.\n- Tell the user, naming the issue but not detailing it, and point them at the\n  responsible-disclosure process.\n- Never put it in a public commit, PR, comment, or issue, and never draft a public response\n  to it. A public reference to an unpatched vulnerability is a leak, not a fix.\n- Never quote exploit detail into a report, log, or plan file.\n- Never label it, in this tracker or any other public place, and never record it in a file.\n  The chat is the only place it appears. With no label, it stays untriaged and gets flagged\n  again on the next run, which is the reminder that it's still open.\n\n## Rules\n\n- Read-only against the codebase. The only file this skill writes is the `## Issue tracker`\n  section of `CLAUDE.md`, when it or one of its lines is missing. The tracker is written to only as the next rule\n  allows.\n- Never post a comment, apply a label, or close an issue without the user picking that action\n  explicitly — issue by issue, or as a stated \"apply all.\" A draft is the default output, not\n  an action.\n- If the tracker is unreachable (auth, network, wrong location), stop and say so. Don't triage\n  a stale cached copy or guess at an issue's content.\n\n## Hand off\n\nA triaged issue that turns into real work goes to `issue-resolve` next, with its ID. That\npicks `quick-fix` or the pipeline, and drafts the tracker reply once the fix lands. A\n`state\u002Fneeds-investigation` issue can go to `investigate-bug` first instead, to prove it\nbefore anyone commits to a fix. An issue with a linked open PR goes to `pr-review` with that\nPR number instead — name the skill explicitly, never a generic \"review the PR(s)\", since a\nbusy install can drop skill descriptions from the listing and a bare mention won't route. To\nwork the whole tracker through to ready, `issue-backlog` runs triage, decisions, and\ninvestigations in order. This skill stops at \"here's what I'd do,\" it doesn't start building.\n",{"name":5,"description":7,"argument-hint":24,"user-invocable":25},"[issue ID | ID range | decisions | leave blank for whole tracker]",true,"main","a2f9c28bc56802b4e6db01e03972e497b30ad001","plugins\u002Fumbraco-devkit\u002Fskills\u002Fissue-triage",1,"https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit","https:\u002F\u002Fgithub.com\u002Fmattbrailsford","https:\u002F\u002Fmattbrailsford.dev","AI Staff Engineer @umbraco ","2026-11-12 18:33:53",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":25},[37,40,43],{"tool":38,"command":39},"npx","npx skills add mattbrailsford\u002Fumbraco-devkit --skill issue-triage",{"tool":41,"command":42},"bunx","bunx skills add mattbrailsford\u002Fumbraco-devkit --skill issue-triage",{"tool":44,"command":45},"pnpm","pnpm dlx skills add mattbrailsford\u002Fumbraco-devkit --skill issue-triage","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fissue-triage\u002FSKILL.md",[],[49,56,63,70,77,84],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":53,"categories":54,"keywords":55,"publishedAt":19},"01M7CYGRPVNP8H9YNHGMZ6CP3C","user-stories","Writes, splits and refines user stories with Given\u002FWhen\u002FThen acceptance criteria in STORIES.md, feeding bdd-specs. Use when turning a feature request or a vague ask into a testable backlog entry.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fuser-stories\u002FSKILL.md",[],[],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7CYGJB26N83DXKMDEKY2X48","umbraco-security","What Umbraco provides for keeping a package secure, on the server and in the backoffice: Management API authorization and policies, settings that hold secrets, endpoints outside the Management API, sanitizeHTML and escapeHTML, DOMPurify rules for custom elements, and umbHttpClient instead of handling tokens in the browser. Use when writing or reviewing a Management API controller, a secret setting, a custom endpoint, a backoffice template that uses unsafeHTML, or frontend code that calls an API.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fumbraco-security\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7CYGBEZ2Q4AXDDNADWW4Q9H","umbraco-performance","Keeping a package fast on the sites that install it: the published cache over the services layer, tree traversal, caching lookups and clearing them, data type config and converter cache levels, output cache eviction, light hot paths, and backoffice screens that don't over-fetch. Use when writing, changing or reviewing code that runs per request, per save, per indexed item or at startup (property value converters, notification handlers, indexing), code that loops over every content item (a full rebuild, an import, a sync), or a list or search screen. Not for a site's own templates.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fumbraco-performance\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7CYG59CZN1K2KERB0SSDGSY","umbraco-package-conventions","Backend conventions for an Umbraco package: async naming, internal by default, Attempt-based results for routine failures, public-API backwards compatibility, and where extension methods live. Use when naming a method, choosing visibility, shaping a service's result, changing a public signature (including adding a dependency to a public class's constructor), or organizing extension methods. For extension mechanisms, see `umbraco-extensibility`.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fumbraco-package-conventions\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7CYG0J10QH8J2GS825WTTET","umbraco-marketplace","Gets a finished package listed on the Umbraco Marketplace: the NuGet tag, the dependency Marketplace reads version support from, the v14+ compatibility gotcha, and umbraco-marketplace.json. Use when a package is ready to publish, or when reviewing its packaging.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fumbraco-marketplace\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7CYFT7MSYW4GRDSQRDPG020","umbraco-extensibility","Umbraco's extension mechanisms (Composers, collection builders, notification handlers, attribute discovery) and a catalogue of backend CMS extension points: property value converters, Management API endpoints, content finders, health checks, search indexing, package migrations, background jobs, and more. Use when making a feature extensible or choosing which extension point fits. Not for property editor UI, dashboards, sections or trees, which the official Backoffice Skills plugin covers.","https:\u002F\u002Fgithub.com\u002Fmattbrailsford\u002Fumbraco-devkit\u002Fblob\u002Fmain\u002Fplugins\u002Fumbraco-devkit\u002Fskills\u002Fumbraco-extensibility\u002FSKILL.md",[],[]]