[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M782JA3VC2Z7SE03S4PDEYJ1":3,"skill-same:01M782JA3VC2Z7SE03S4PDEYJ1":53},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M782JA3VC2Z7SE03S4PDEYJ1","complete",null,"Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval before squash merge. Use for \u002Fcomplete or requests to finish, wrap up, merge, or close the current work item.","bradtraversy\u002Fskillpass","bradtraversy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F5550850?v=4","01M782H8BT6A97A17XDFHDABWZ","Brad Traversy","01M782H8CDJZW3HYDHX23FA9VF",26,"MIT","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcomplete\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: complete\ndescription: Complete a finished feature, fix, or rollback by running final gates, archiving its spec, updating plans, creating the work commit, and requesting approval before squash merge. Use for \u002Fcomplete or requests to finish, wrap up, merge, or close the current work item.\n---\n\n# complete - log the finished work, make the work commit, and merge\n\n**Context reuse:** Reuse any required file already loaded in project instructions or the current session. Read it again only if absent, changed, or exact current bytes or line references are needed.\n\n**First action:** Before project inspection, preflight, or any other tool call,\npublish `running` to `blueprint\u002F.state\u002Frun.json` using the dashboard activity\ncontract in `AGENTS.md`.\n\nWhere this sits in the workflow:\n\n    \u002Ffeature, \u002Ffix, or \u002Frollback  ->  \u002Fimplement  ->  [complete]  ->  next\n    (the spec)                         (build it)      (commit + merge + log)\n\n`\u002Fimplement` built the feature, fix, or rollback on its branch, with optional per-step commit\ncheckpoints. This skill closes it out: it logs the work, makes the single\nwork-level commit, and squash-merges. Run it only when the work is done,\nreviewed, and the documented `Verify` command, or the fallback build and tests,\npasses.\n\n## Before you start\n\nRead `blueprint\u002Fconfig.json`. A missing file means the built-in defaults apply.\nIf the file exists but is invalid, stop and point the user to `\u002Fdoctor`.\nConfiguration can strengthen or shape the completion gates, but it never grants\npermission to commit, merge, push, deploy, publish, or take destructive action.\n\nBefore requiring a real active spec, check for pending completion using\n`reference\u002Fcompletion-recovery.md`. A matching archive may mean archival was\ninterrupted, the work commit awaits merge, or the merge already finished. Follow\nthat phase instead of restarting logging. Missing or ambiguous evidence stops\nwith concrete recovery steps; a reset stub never authorizes selecting new work.\nRecovery uses archive and Git proof, never dashboard activity as authority.\n\nFor a normal completion with no recovery in progress, continue below.\n\nConfirm the work is actually finished: `blueprint\u002Fcontext\u002Fcurrent-feature.md`\nholds a real spec, its steps are built on a branch, and `Verify`, or the fallback\nbuild and tests, passes. Apply the configured regular quality gates below before\nlogging or committing. Uncommitted step work is expected because per-step\ncheckpoints are optional; this skill commits it. Don't require the steps to be\npre-committed.\n\nRead `blueprint\u002Fcontext\u002Freview.md` when present. A missing file on an older\ninstall means no independent review has been requested. A pending,\nchanges-requested, malformed, or stale record is always a blocker because the\nuser already initiated that gate, even when its configured policy is `manual`.\nComplete must route a pending or missing required review through the configured\nexecution flow below rather than treating the manual handoff as unconditional.\n\n## Configured regular quality gates\n\nUse `qualityGates.regular` for this work item:\n\n- **Audit:** `manual` runs only when the user explicitly requests `\u002Faudit`;\n  `when-sensitive` runs for authentication, authorization, payments, secrets,\n  personal or user data, migrations, destructive operations, external side\n  effects, security boundaries, or unusually broad changes; `always` runs for\n  every work item.\n- **Independent review:** `manual` runs only when the user explicitly requests\n  `\u002Faudit independent current`; `when-sensitive` requires it for the same\n  sensitive categories as Audit; `always` requires it for every work item. An\n  independent review includes a full current audit, so it satisfies a selected\n  Audit gate instead of repeating the audit in the builder session.\n- **Check:** `manual` runs only when explicitly requested; `when-behavioral` runs\n  when a done-when needs observed runtime behavior such as a click, request, CLI\n  command, download, background job, or multi-screen flow; `always` runs for\n  every work item.\n- **Try guide (`qualityGates.regular.tryGuide`):** use `\u002Fcheck guide`.\n  `manual` runs only when explicitly requested;\n  `when-user-facing` generates a guide when the change affects UI, navigation,\n  copy, a public API or CLI, output, or another workflow a person directly uses;\n  `always` generates one for every work item.\n\nApply automatic gates in this order: `\u002Fcheck`, review, then `\u002Fcheck guide`. When\nindependent review is selected, follow the independent execution flow below and\ncontinue only after a fresh reviewer writes a current passing receipt. Otherwise\nrun `\u002Faudit current` when Audit is selected.\nReuse adequate evidence produced during the current work item instead of\nrepeating it. A required gate that cannot run is a blocker. `\u002Fcheck guide` only\ngenerates instructions for human review; never claim the user performed them. P0 and P1\nfinding blockers remain enforced regardless of these settings.\n\n### Independent review execution\n\nAfter final Verify and required Check pass, set the active spec to `verified`.\nIf a selected or previously initiated independent review does not already have a\ncurrent passing receipt:\n\n1. Use an existing current pending request and its immutable target when one is\n   present. Otherwise show the exact product\u002Ftest checkpoint candidate and\n   verified spec. Include the spec when tracked; an intentionally ignored spec\n   uses Audit's local `Spec snapshot` contract without changing visibility.\n   Obtain explicit commit approval when the exact checkpoint does not already\n   exist, then create or use it under the normal Git rules. Configuration,\n   including `review.independentExecution: \"automatic\"`, never grants permission\n   to commit. A pending request without `Requested execution` is legacy and\n   manual-only; never add execution fields or run a subagent against it.\n   A local-spec-only revision may reuse the same approved product HEAD after\n   normal spec and verification gates, with a new snapshot\u002Frequest and full\n   fresh review. Do not create an empty commit for ignored spec changes.\n2. Prepare Phase A of `\u002Faudit independent current` when no current request\n   exists. Record `Requested execution` from `review.independentExecution`.\n3. For requested `automatic`, start the generic isolated current-runtime child\n   from the installed project-local Audit skill, wait, and validate the normal\n   receipt. Freeze parent product, test, spec, and config changes while it runs.\n4. For requested `manual`, or when automatic isolation, identity, model,\n   completion, or access to the same local spec\u002Fsnapshot is unavailable,\n   preserve the pending request, set activity to `ready`, and stop with the\n   manual fresh-session handoff.\n5. Continue Complete only with a current passing receipt whose requested and\n   actual execution fields match the allowed review contract. Never self-review\n   or silently skip the gate.\n\n## Step 0 - final safety pass\n\nBefore logging or committing, run a short safety pass and report blockers only:\n\n- active spec exists and the work is not being completed from `main` or `master`\n- the branch name uses the configured feature, fix, or rollback prefix\n- changed files are tied to the active spec, with no unrelated dirty work mixed\n  in (dirty `blueprint\u002Fcontext\u002Ffindings.md` and\n  `blueprint\u002Fcontext\u002Freview.md` are expected review evidence)\n- the exact `Verify` command from `AGENTS.md` passed in this session, when one is\n  declared; otherwise the build passed, and tests passed when the project has a\n  declared test command and the change touched logic\n- any check required by `qualityGates.regular` has evidence, and there is a clear\n  manual try path\n- with `verification.logicTests: \"required\"`, logic changes have a configured\n  runner and passing focused tests; otherwise completion is blocked and `\u002Ftests`\n  is the next setup step\n- with `verification.uiEvidence: \"required\"`, UI done-whens have direct browser\n  evidence, including screenshots and relevant console and network checks\n- any audit or try guide required by `qualityGates.regular` ran before completion\n- a selected independent-review gate has a `passed` receipt whose target equals\n  `HEAD`, whose base ref still produces the recorded merge base, whose exact\n  spec SHA-256 still matches, whose reviewer adapter and selected model match\n  the request, whose required Check result passed, whose receipt sections are\n  non-empty, and whose target has no later changes except the review and\n  findings files. Apply the same checks to any explicit receipt even when the\n  configured policy is `manual`. Any mismatch is stale and blocks completion.\n  When `Spec snapshot` is present, also require its exact local bytes, path,\n  visibility, and Git conditions from Audit's reference contract to remain valid.\n- when a passing independent receipt exists, the active spec is already\n  `verified` and remains byte-for-byte unchanged through archival\n- if workflow files changed, `.agents` and `.claude` stayed in sync where both\n  adapters exist\n- no P0 or P1 finding in `blueprint\u002Fcontext\u002Ffindings.md` is `open` or `fixed`.\n  `fixed` still blocks on purpose: the repair exists but no review has looked at\n  it - run `\u002Faudit` to close it. While the current spec is active, append the\n  repair as a new checklist step and run `\u002Fimplement`; do not start a separate\n  `\u002Ffix`. After the repair passes its focused check and the finding is marked\n  `fixed`, run `\u002Faudit current` to re-review the updated code, close the finding,\n  and then retry `\u002Fcomplete`. The only waivers are `accepted` (the user's\n  explicit decision in the current chat, reason recorded; never set it for\n  them) or `invalid` (an `\u002Faudit` re-examination verdict with recorded\n  evidence, or the user's explicit call). A missing ledger file means no\n  findings.\n\nDo not claim \"passed\", \"verified\", or \"working\" without naming the command,\nroute, screenshot, or output that proves it. Stop before Step 1 if required\nevidence is missing.\n\nAfter this safety pass succeeds, set the active spec's `**Status:**` to\n`verified` before archiving it when no independent receipt exists. With a\npassing independent receipt, it must already be `verified`; do not rewrite it\nafter review. Rerun the required final checks either way because `\u002Fcomplete`\nowns the final safety pass.\n\n## Step 1 - log the work\n\nFollow `reference\u002Fcompletion-recovery.md` to capture the source tree and compact\narchive annotation before any logging edits. Preserve the exact verified spec\nprefix, its UTF-8 byte length and SHA-256, branch, original HEAD, and local base.\nRecord the reference's narrow `absentOptional` proof before creating any optional\nfindings\u002Freview stub on an older installation; tree absence alone is insufficient.\nPrepare the entire archive, including the sections below and any generated try\nguide, before placing it at its absent destination. An existing matching archive\nenters recovery; never overwrite it or append duplicate sections.\n\nCheck whether the spec is a feature, fix, or rollback. A fix is marked\n`Type: Fix` and has no build-plan number. A rollback is marked `Type: Rollback`\nand records the exact target feature, archive, commit, and parent.\n\n- **Feature** - use the verified spec's frozen `**Build attempt:**` and stable ID,\n  following `..\u002Ffeature\u002Freference\u002Fbuild-history.md`. Attempt 1 keeps\n  `blueprint\u002Fhistory\u002Ffeatures\u002FNN-name.md`; N > 1 uses `NN-name--build-N.md`.\n  Validate the attempt against prior immutable builds and completed reversals;\n  never increment it at completion or resume. For an older active spec without\n  the field, derive 1 only with no prior builds, or derive the next attempt only\n  from unambiguous prior builds and their proven completed reversals. Keep those\n  reviewed spec bytes and the existing branch unchanged; otherwise stop. An\n  existing matching archive and its annotation freeze the recovery destination.\n  Check the feature off in `blueprint\u002Fbuild-plan.md`\n  (and its parent item once all sub-items are checked). Then recompute the\n  overview fingerprint using `\u002Foverview`'s checkbox-normalized hash contract and\n  replace only the existing `blueprint:source-hash` value. Do not regenerate or\n  rewrite the overview body. This keeps completion progress from looking like\n  plan drift and migrates older exact-byte fingerprints.\n- **Fix** - archive it to `blueprint\u002Fhistory\u002Ffixes\u002Fname.md`. A fix isn't a build-plan item, so\n  there's nothing to check off.\n- **Rollback** - archive it to\n  `blueprint\u002Fhistory\u002Frollbacks\u002FYYYY-MM-DD-\u003Cexact-target-archive-stem>.md`, using the\n  spec's exact `Target archive` filename without `.md`, including any build suffix.\n  Preserve the original completed feature archive. Create\n  `blueprint\u002Fhistory\u002Frollbacks\u002F` first if an\n  older Blueprint installation does not have it yet. Uncheck the exact target item in\n  `blueprint\u002Fbuild-plan.md` and its parent when applicable, then append a concise\n  note to the target line with the rollback date and archive path. Keep the\n  feature number stable. If the user later decides the feature is permanently\n  abandoned rather than pending rebuild, that roadmap decision is a separate\n  plan edit.\n\n**Archive resolved findings.** If `blueprint\u002Fcontext\u002Ffindings.md` holds any\nfindings, include a `## Findings` section in the prepared archive with\nevery `closed`, `accepted`, or `invalid` entry at its final status (`accepted`\nentries keep their recorded reason). Prefix feature IDs with the normalized stable\nID and, for N > 1, `-build-N`: feature 12's first `F-03` becomes `12\u002FF-03`, while\nattempt 2 becomes `12-build-2\u002FF-03`. Derive N from the verified spec\u002Fhistory proof,\nnever arbitrary filename text. Fixes and rollbacks use their archive filename as\nthe prefix. An entry carried forward from earlier work archives with the item that\nresolved it; its **Found** line preserves\nwhere it came from. Only `closed`, `accepted`, and `invalid` entries are\nresolved for archival. A `fixed` entry is not resolved at any severity: never\nappend it to the archive or remove it from the live ledger.\n\n**Archive independent review.** When a current `passed` receipt exists, include\na `## Independent review` section in the prepared archive with the receipt fields,\ncommands, safe evidence references, findings, and remaining risk from\n`blueprint\u002Fcontext\u002Freview.md`. Preserve the full target and base SHAs, spec\nhash, the original `Spec snapshot` field when present, base ref, builder adapter\nand model, requested reviewer, model, and execution, actual reviewer adapter,\nmodel, and execution, Check result, fresh-context declaration, and review time.\nDo not archive a stale, pending,\nchanges-requested, or malformed record.\n\nValidate and place the fully assembled archive first. Complete only the exact\nplan\u002Foverview changes above and any approved consumed-prototype cleanup below.\nRead the archive back and confirm that its spec, findings, and review match their\ninputs before resetting live evidence. Reset the active spec last.\n\n**Discard consumed prototypes.** If this feature built the look from `prototypes\u002F`\n- its Design reference pointed there and an early step ported `prototypes\u002Ftheme.css`\ninto the app - delete the `prototypes\u002F` folder now. The tokens live in the real\nstylesheet and the HTML mockups were always throwaway; fold the deletion into this\nfeature's commit. Skip this if the feature didn't consume prototypes.\n\nThen remove only the archived entries from the ledger. Entries with `open`,\n`fixed`, or `unverified` status stay in the ledger with their IDs so they are\nnever silently dropped. A fixed P2\u002FP3 finding does not block completion, but it\nmust remain verbatim for a later `\u002Faudit` re-review. When no `open`, `fixed`, or\n`unverified` entries remain, reset the ledger to exactly this stub, and create it the\nsame way if the file is missing (an older install):\n\n    # Findings\n\n    > **Generated file.** The findings ledger: review findings raised by `\u002Faudit`\n    > against the work in progress, each with a durable ID, severity (P0-P3), and\n    > status. `\u002Fimplement` marks repaired findings `fixed`, a later `\u002Faudit` pass\n    > moves them to `closed`, and `\u002Fcomplete` refuses to merge while any P0 or P1\n    > finding is `open` or `fixed`, then archives resolved findings with the work\n    > and resets this file.\n\n    _No findings recorded. `\u002Faudit` appends findings here when it finds them._\n\nThen reset `blueprint\u002Fcontext\u002Freview.md` to exactly this stub, creating it when\nan older installation does not have it:\n\n    # Independent Review\n\n    > **Generated file.** Holds the active independent-review request or latest\n    > receipt for the current work item. `\u002Faudit independent current` prepares a\n    > handoff against an approved checkpoint, a fresh reviewer context completes it,\n    > and `\u002Fcomplete` refuses stale, pending, or changes-requested review state.\n\n    _No independent review requested. Run `\u002Faudit independent current` to prepare one._\n\nKeep every unresolved entry in the ledger. Do not replace it with the empty stub\nwhile it still contains any open, fixed, or unverified finding. After archiving\nresolved findings, replace `blueprint\u002Fcontext\u002Fcurrent-feature.md` with\nthe canonical stub below. Do not paraphrase it or substitute an abbreviated \"no\nwork\" stub. Before committing, read the file and confirm it exactly matches:\n\n    # Current Feature\n\n    > **Generated file.** Holds the one feature, fix, or rollback being built right now. Run\n    > `\u002Ffeature \u003Cnumber-or-name>` to spec a build-plan feature, or `\u002Ffix \"\u003Cbug>\"` for\n    > an ad-hoc fix. Use `\u002Frollback \u003Ccompleted-feature>` to plan a safe reversal.\n    > Build one thing at a time; `\u002Fcomplete` archives it under\n    > `blueprint\u002Fhistory\u002F` and resets this file.\n\n    _Nothing in progress. Run `\u002Ffeature`, `\u002Ffix`, or `\u002Frollback` to start._\n\nWhen no open, fixed, or unverified ledger entries remain, confirm\n`blueprint\u002Fcontext\u002Ffindings.md` exactly matches the canonical Findings stub\nabove. Otherwise, preserve the remaining entries without rewriting them.\nConfirm `blueprint\u002Fcontext\u002Freview.md` exactly matches the canonical Independent\nReview stub above.\n\nDon't commit yet; the next step makes one work commit covering the code and these\ndocumentation changes. The archive is the build history.\n\n## Step 2 - make the work commit\n\nShow the complete product and logging diff with the proposed commit message,\nthen obtain explicit commit approval. Only then stage the reviewed branch work\n(any uncommitted step work plus the Step 1 logging changes) and make one conventional work commit (for example `feat: \u003Cfeature>`,\n`fix: \u003Cname>`, or `revert: roll back \u003Cfeature>`). `Verify`, or the fallback build\nand tests, must pass first.\n\n## Step 3 - merge\n\n1. Confirm the recorded local default branch has not advanced and the final work\n   commit is unchanged. Squash-merge into that default branch only with the user's\n   explicit go-ahead, so\n   the feature lands as one clean commit regardless of how many checkpoints the\n   branch carried.\n2. Verify the resulting default-branch commit, parent, archive, and full tree\n   using `reference\u002Fcompletion-recovery.md`, then perform approved branch cleanup.\n3. Stop and ask whether to push local `main` to its upstream. The merge approval\n   does not count as push approval.\n4. Push main only after a separate explicit yes to push main in the current chat.\n   If the repo has no remote or upstream, say so instead of guessing.\n\nThen point the user at `\u002Ffeature`, `\u002Ffix`, or `\u002Frollback` for the next thing.\n\nFinish with a concise **How to try it** note for the completed work. For a\nrollback, explain how to confirm the removed behavior is gone and name one\nunaffected regression path. If the\nmanual path is more than a couple of steps, tell the user to run `\u002Fcheck guide latest`;\nthat command can read the archived feature after `current-feature.md` is reset.\n\n## Rules\n\n- The work item is the unit of history: one squashed feature, fix, or rollback\n  commit on main, even if the branch carried several checkpoint commits.\n- A rollback preserves the original feature archive and adds a separate rollback\n  archive. Never rewrite history to make the feature look as if it never existed.\n- Don't merge unfinished or failing work. The documented `Verify` command, or\n  the fallback build and tests, must pass first.\n- Never merge while a P0 or P1 finding is `open` or `fixed` in the ledger. The\n  recorded ways past the gate without code are `accepted` (only by the user's\n  explicit decision, with their reason) or `invalid` (only from re-examination\n  evidence or the user's explicit call); both travel into the archive, never a\n  silent drop.\n- Never merge with a required or explicitly initiated independent review that\n  is missing, pending, changes-requested, malformed, or stale. The user may\n  explicitly cancel a manual review before completion, but the agent never\n  resets or waives it on the user's behalf.\n- Merging and pushing are the user's calls: get an explicit yes for the merge,\n  then ask whether to push main. Do not treat merge approval, `\u002Fcomplete`, or\n  \"looks good\" as permission to push.\n- Push main only after a separate explicit yes to push main in the current chat.\n- One item per completion. If a parent feature still has unchecked sub-features,\n  leave the parent unchecked.\n\n## Formatting\n\nFormat the output to match the project's conventions in\n`blueprint\u002Fcontext\u002Fai-interaction.md`: concise, scannable markdown, with lists for\nenumerations and tables for matrices rather than dense paragraphs.\n","# complete - log the finished work, make the work commit, and merge\n\n**Context reuse:** Reuse any required file already loaded in project instructions or the current session. Read it again only if absent, changed, or exact current bytes or line references are needed.\n\n**First action:** Before project inspection, preflight, or any other tool call,\npublish `running` to `blueprint\u002F.state\u002Frun.json` using the dashboard activity\ncontract in `AGENTS.md`.\n\nWhere this sits in the workflow:\n\n    \u002Ffeature, \u002Ffix, or \u002Frollback  ->  \u002Fimplement  ->  [complete]  ->  next\n    (the spec)                         (build it)      (commit + merge + log)\n\n`\u002Fimplement` built the feature, fix, or rollback on its branch, with optional per-step commit\ncheckpoints. This skill closes it out: it logs the work, makes the single\nwork-level commit, and squash-merges. Run it only when the work is done,\nreviewed, and the documented `Verify` command, or the fallback build and tests,\npasses.\n\n## Before you start\n\nRead `blueprint\u002Fconfig.json`. A missing file means the built-in defaults apply.\nIf the file exists but is invalid, stop and point the user to `\u002Fdoctor`.\nConfiguration can strengthen or shape the completion gates, but it never grants\npermission to commit, merge, push, deploy, publish, or take destructive action.\n\nBefore requiring a real active spec, check for pending completion using\n`reference\u002Fcompletion-recovery.md`. A matching archive may mean archival was\ninterrupted, the work commit awaits merge, or the merge already finished. Follow\nthat phase instead of restarting logging. Missing or ambiguous evidence stops\nwith concrete recovery steps; a reset stub never authorizes selecting new work.\nRecovery uses archive and Git proof, never dashboard activity as authority.\n\nFor a normal completion with no recovery in progress, continue below.\n\nConfirm the work is actually finished: `blueprint\u002Fcontext\u002Fcurrent-feature.md`\nholds a real spec, its steps are built on a branch, and `Verify`, or the fallback\nbuild and tests, passes. Apply the configured regular quality gates below before\nlogging or committing. Uncommitted step work is expected because per-step\ncheckpoints are optional; this skill commits it. Don't require the steps to be\npre-committed.\n\nRead `blueprint\u002Fcontext\u002Freview.md` when present. A missing file on an older\ninstall means no independent review has been requested. A pending,\nchanges-requested, malformed, or stale record is always a blocker because the\nuser already initiated that gate, even when its configured policy is `manual`.\nComplete must route a pending or missing required review through the configured\nexecution flow below rather than treating the manual handoff as unconditional.\n\n## Configured regular quality gates\n\nUse `qualityGates.regular` for this work item:\n\n- **Audit:** `manual` runs only when the user explicitly requests `\u002Faudit`;\n  `when-sensitive` runs for authentication, authorization, payments, secrets,\n  personal or user data, migrations, destructive operations, external side\n  effects, security boundaries, or unusually broad changes; `always` runs for\n  every work item.\n- **Independent review:** `manual` runs only when the user explicitly requests\n  `\u002Faudit independent current`; `when-sensitive` requires it for the same\n  sensitive categories as Audit; `always` requires it for every work item. An\n  independent review includes a full current audit, so it satisfies a selected\n  Audit gate instead of repeating the audit in the builder session.\n- **Check:** `manual` runs only when explicitly requested; `when-behavioral` runs\n  when a done-when needs observed runtime behavior such as a click, request, CLI\n  command, download, background job, or multi-screen flow; `always` runs for\n  every work item.\n- **Try guide (`qualityGates.regular.tryGuide`):** use `\u002Fcheck guide`.\n  `manual` runs only when explicitly requested;\n  `when-user-facing` generates a guide when the change affects UI, navigation,\n  copy, a public API or CLI, output, or another workflow a person directly uses;\n  `always` generates one for every work item.\n\nApply automatic gates in this order: `\u002Fcheck`, review, then `\u002Fcheck guide`. When\nindependent review is selected, follow the independent execution flow below and\ncontinue only after a fresh reviewer writes a current passing receipt. Otherwise\nrun `\u002Faudit current` when Audit is selected.\nReuse adequate evidence produced during the current work item instead of\nrepeating it. A required gate that cannot run is a blocker. `\u002Fcheck guide` only\ngenerates instructions for human review; never claim the user performed them. P0 and P1\nfinding blockers remain enforced regardless of these settings.\n\n### Independent review execution\n\nAfter final Verify and required Check pass, set the active spec to `verified`.\nIf a selected or previously initiated independent review does not already have a\ncurrent passing receipt:\n\n1. Use an existing current pending request and its immutable target when one is\n   present. Otherwise show the exact product\u002Ftest checkpoint candidate and\n   verified spec. Include the spec when tracked; an intentionally ignored spec\n   uses Audit's local `Spec snapshot` contract without changing visibility.\n   Obtain explicit commit approval when the exact checkpoint does not already\n   exist, then create or use it under the normal Git rules. Configuration,\n   including `review.independentExecution: \"automatic\"`, never grants permission\n   to commit. A pending request without `Requested execution` is legacy and\n   manual-only; never add execution fields or run a subagent against it.\n   A local-spec-only revision may reuse the same approved product HEAD after\n   normal spec and verification gates, with a new snapshot\u002Frequest and full\n   fresh review. Do not create an empty commit for ignored spec changes.\n2. Prepare Phase A of `\u002Faudit independent current` when no current request\n   exists. Record `Requested execution` from `review.independentExecution`.\n3. For requested `automatic`, start the generic isolated current-runtime child\n   from the installed project-local Audit skill, wait, and validate the normal\n   receipt. Freeze parent product, test, spec, and config changes while it runs.\n4. For requested `manual`, or when automatic isolation, identity, model,\n   completion, or access to the same local spec\u002Fsnapshot is unavailable,\n   preserve the pending request, set activity to `ready`, and stop with the\n   manual fresh-session handoff.\n5. Continue Complete only with a current passing receipt whose requested and\n   actual execution fields match the allowed review contract. Never self-review\n   or silently skip the gate.\n\n## Step 0 - final safety pass\n\nBefore logging or committing, run a short safety pass and report blockers only:\n\n- active spec exists and the work is not being completed from `main` or `master`\n- the branch name uses the configured feature, fix, or rollback prefix\n- changed files are tied to the active spec, with no unrelated dirty work mixed\n  in (dirty `blueprint\u002Fcontext\u002Ffindings.md` and\n  `blueprint\u002Fcontext\u002Freview.md` are expected review evidence)\n- the exact `Verify` command from `AGENTS.md` passed in this session, when one is\n  declared; otherwise the build passed, and tests passed when the project has a\n  declared test command and the change touched logic\n- any check required by `qualityGates.regular` has evidence, and there is a clear\n  manual try path\n- with `verification.logicTests: \"required\"`, logic changes have a configured\n  runner and passing focused tests; otherwise completion is blocked and `\u002Ftests`\n  is the next setup step\n- with `verification.uiEvidence: \"required\"`, UI done-whens have direct browser\n  evidence, including screenshots and relevant console and network checks\n- any audit or try guide required by `qualityGates.regular` ran before completion\n- a selected independent-review gate has a `passed` receipt whose target equals\n  `HEAD`, whose base ref still produces the recorded merge base, whose exact\n  spec SHA-256 still matches, whose reviewer adapter and selected model match\n  the request, whose required Check result passed, whose receipt sections are\n  non-empty, and whose target has no later changes except the review and\n  findings files. Apply the same checks to any explicit receipt even when the\n  configured policy is `manual`. Any mismatch is stale and blocks completion.\n  When `Spec snapshot` is present, also require its exact local bytes, path,\n  visibility, and Git conditions from Audit's reference contract to remain valid.\n- when a passing independent receipt exists, the active spec is already\n  `verified` and remains byte-for-byte unchanged through archival\n- if workflow files changed, `.agents` and `.claude` stayed in sync where both\n  adapters exist\n- no P0 or P1 finding in `blueprint\u002Fcontext\u002Ffindings.md` is `open` or `fixed`.\n  `fixed` still blocks on purpose: the repair exists but no review has looked at\n  it - run `\u002Faudit` to close it. While the current spec is active, append the\n  repair as a new checklist step and run `\u002Fimplement`; do not start a separate\n  `\u002Ffix`. After the repair passes its focused check and the finding is marked\n  `fixed`, run `\u002Faudit current` to re-review the updated code, close the finding,\n  and then retry `\u002Fcomplete`. The only waivers are `accepted` (the user's\n  explicit decision in the current chat, reason recorded; never set it for\n  them) or `invalid` (an `\u002Faudit` re-examination verdict with recorded\n  evidence, or the user's explicit call). A missing ledger file means no\n  findings.\n\nDo not claim \"passed\", \"verified\", or \"working\" without naming the command,\nroute, screenshot, or output that proves it. Stop before Step 1 if required\nevidence is missing.\n\nAfter this safety pass succeeds, set the active spec's `**Status:**` to\n`verified` before archiving it when no independent receipt exists. With a\npassing independent receipt, it must already be `verified`; do not rewrite it\nafter review. Rerun the required final checks either way because `\u002Fcomplete`\nowns the final safety pass.\n\n## Step 1 - log the work\n\nFollow `reference\u002Fcompletion-recovery.md` to capture the source tree and compact\narchive annotation before any logging edits. Preserve the exact verified spec\nprefix, its UTF-8 byte length and SHA-256, branch, original HEAD, and local base.\nRecord the reference's narrow `absentOptional` proof before creating any optional\nfindings\u002Freview stub on an older installation; tree absence alone is insufficient.\nPrepare the entire archive, including the sections below and any generated try\nguide, before placing it at its absent destination. An existing matching archive\nenters recovery; never overwrite it or append duplicate sections.\n\nCheck whether the spec is a feature, fix, or rollback. A fix is marked\n`Type: Fix` and has no build-plan number. A rollback is marked `Type: Rollback`\nand records the exact target feature, archive, commit, and parent.\n\n- **Feature** - use the verified spec's frozen `**Build attempt:**` and stable ID,\n  following `..\u002Ffeature\u002Freference\u002Fbuild-history.md`. Attempt 1 keeps\n  `blueprint\u002Fhistory\u002Ffeatures\u002FNN-name.md`; N > 1 uses `NN-name--build-N.md`.\n  Validate the attempt against prior immutable builds and completed reversals;\n  never increment it at completion or resume. For an older active spec without\n  the field, derive 1 only with no prior builds, or derive the next attempt only\n  from unambiguous prior builds and their proven completed reversals. Keep those\n  reviewed spec bytes and the existing branch unchanged; otherwise stop. An\n  existing matching archive and its annotation freeze the recovery destination.\n  Check the feature off in `blueprint\u002Fbuild-plan.md`\n  (and its parent item once all sub-items are checked). Then recompute the\n  overview fingerprint using `\u002Foverview`'s checkbox-normalized hash contract and\n  replace only the existing `blueprint:source-hash` value. Do not regenerate or\n  rewrite the overview body. This keeps completion progress from looking like\n  plan drift and migrates older exact-byte fingerprints.\n- **Fix** - archive it to `blueprint\u002Fhistory\u002Ffixes\u002Fname.md`. A fix isn't a build-plan item, so\n  there's nothing to check off.\n- **Rollback** - archive it to\n  `blueprint\u002Fhistory\u002Frollbacks\u002FYYYY-MM-DD-\u003Cexact-target-archive-stem>.md`, using the\n  spec's exact `Target archive` filename without `.md`, including any build suffix.\n  Preserve the original completed feature archive. Create\n  `blueprint\u002Fhistory\u002Frollbacks\u002F` first if an\n  older Blueprint installation does not have it yet. Uncheck the exact target item in\n  `blueprint\u002Fbuild-plan.md` and its parent when applicable, then append a concise\n  note to the target line with the rollback date and archive path. Keep the\n  feature number stable. If the user later decides the feature is permanently\n  abandoned rather than pending rebuild, that roadmap decision is a separate\n  plan edit.\n\n**Archive resolved findings.** If `blueprint\u002Fcontext\u002Ffindings.md` holds any\nfindings, include a `## Findings` section in the prepared archive with\nevery `closed`, `accepted`, or `invalid` entry at its final status (`accepted`\nentries keep their recorded reason). Prefix feature IDs with the normalized stable\nID and, for N > 1, `-build-N`: feature 12's first `F-03` becomes `12\u002FF-03`, while\nattempt 2 becomes `12-build-2\u002FF-03`. Derive N from the verified spec\u002Fhistory proof,\nnever arbitrary filename text. Fixes and rollbacks use their archive filename as\nthe prefix. An entry carried forward from earlier work archives with the item that\nresolved it; its **Found** line preserves\nwhere it came from. Only `closed`, `accepted`, and `invalid` entries are\nresolved for archival. A `fixed` entry is not resolved at any severity: never\nappend it to the archive or remove it from the live ledger.\n\n**Archive independent review.** When a current `passed` receipt exists, include\na `## Independent review` section in the prepared archive with the receipt fields,\ncommands, safe evidence references, findings, and remaining risk from\n`blueprint\u002Fcontext\u002Freview.md`. Preserve the full target and base SHAs, spec\nhash, the original `Spec snapshot` field when present, base ref, builder adapter\nand model, requested reviewer, model, and execution, actual reviewer adapter,\nmodel, and execution, Check result, fresh-context declaration, and review time.\nDo not archive a stale, pending,\nchanges-requested, or malformed record.\n\nValidate and place the fully assembled archive first. Complete only the exact\nplan\u002Foverview changes above and any approved consumed-prototype cleanup below.\nRead the archive back and confirm that its spec, findings, and review match their\ninputs before resetting live evidence. Reset the active spec last.\n\n**Discard consumed prototypes.** If this feature built the look from `prototypes\u002F`\n- its Design reference pointed there and an early step ported `prototypes\u002Ftheme.css`\ninto the app - delete the `prototypes\u002F` folder now. The tokens live in the real\nstylesheet and the HTML mockups were always throwaway; fold the deletion into this\nfeature's commit. Skip this if the feature didn't consume prototypes.\n\nThen remove only the archived entries from the ledger. Entries with `open`,\n`fixed`, or `unverified` status stay in the ledger with their IDs so they are\nnever silently dropped. A fixed P2\u002FP3 finding does not block completion, but it\nmust remain verbatim for a later `\u002Faudit` re-review. When no `open`, `fixed`, or\n`unverified` entries remain, reset the ledger to exactly this stub, and create it the\nsame way if the file is missing (an older install):\n\n    # Findings\n\n    > **Generated file.** The findings ledger: review findings raised by `\u002Faudit`\n    > against the work in progress, each with a durable ID, severity (P0-P3), and\n    > status. `\u002Fimplement` marks repaired findings `fixed`, a later `\u002Faudit` pass\n    > moves them to `closed`, and `\u002Fcomplete` refuses to merge while any P0 or P1\n    > finding is `open` or `fixed`, then archives resolved findings with the work\n    > and resets this file.\n\n    _No findings recorded. `\u002Faudit` appends findings here when it finds them._\n\nThen reset `blueprint\u002Fcontext\u002Freview.md` to exactly this stub, creating it when\nan older installation does not have it:\n\n    # Independent Review\n\n    > **Generated file.** Holds the active independent-review request or latest\n    > receipt for the current work item. `\u002Faudit independent current` prepares a\n    > handoff against an approved checkpoint, a fresh reviewer context completes it,\n    > and `\u002Fcomplete` refuses stale, pending, or changes-requested review state.\n\n    _No independent review requested. Run `\u002Faudit independent current` to prepare one._\n\nKeep every unresolved entry in the ledger. Do not replace it with the empty stub\nwhile it still contains any open, fixed, or unverified finding. After archiving\nresolved findings, replace `blueprint\u002Fcontext\u002Fcurrent-feature.md` with\nthe canonical stub below. Do not paraphrase it or substitute an abbreviated \"no\nwork\" stub. Before committing, read the file and confirm it exactly matches:\n\n    # Current Feature\n\n    > **Generated file.** Holds the one feature, fix, or rollback being built right now. Run\n    > `\u002Ffeature \u003Cnumber-or-name>` to spec a build-plan feature, or `\u002Ffix \"\u003Cbug>\"` for\n    > an ad-hoc fix. Use `\u002Frollback \u003Ccompleted-feature>` to plan a safe reversal.\n    > Build one thing at a time; `\u002Fcomplete` archives it under\n    > `blueprint\u002Fhistory\u002F` and resets this file.\n\n    _Nothing in progress. Run `\u002Ffeature`, `\u002Ffix`, or `\u002Frollback` to start._\n\nWhen no open, fixed, or unverified ledger entries remain, confirm\n`blueprint\u002Fcontext\u002Ffindings.md` exactly matches the canonical Findings stub\nabove. Otherwise, preserve the remaining entries without rewriting them.\nConfirm `blueprint\u002Fcontext\u002Freview.md` exactly matches the canonical Independent\nReview stub above.\n\nDon't commit yet; the next step makes one work commit covering the code and these\ndocumentation changes. The archive is the build history.\n\n## Step 2 - make the work commit\n\nShow the complete product and logging diff with the proposed commit message,\nthen obtain explicit commit approval. Only then stage the reviewed branch work\n(any uncommitted step work plus the Step 1 logging changes) and make one conventional work commit (for example `feat: \u003Cfeature>`,\n`fix: \u003Cname>`, or `revert: roll back \u003Cfeature>`). `Verify`, or the fallback build\nand tests, must pass first.\n\n## Step 3 - merge\n\n1. Confirm the recorded local default branch has not advanced and the final work\n   commit is unchanged. Squash-merge into that default branch only with the user's\n   explicit go-ahead, so\n   the feature lands as one clean commit regardless of how many checkpoints the\n   branch carried.\n2. Verify the resulting default-branch commit, parent, archive, and full tree\n   using `reference\u002Fcompletion-recovery.md`, then perform approved branch cleanup.\n3. Stop and ask whether to push local `main` to its upstream. The merge approval\n   does not count as push approval.\n4. Push main only after a separate explicit yes to push main in the current chat.\n   If the repo has no remote or upstream, say so instead of guessing.\n\nThen point the user at `\u002Ffeature`, `\u002Ffix`, or `\u002Frollback` for the next thing.\n\nFinish with a concise **How to try it** note for the completed work. For a\nrollback, explain how to confirm the removed behavior is gone and name one\nunaffected regression path. If the\nmanual path is more than a couple of steps, tell the user to run `\u002Fcheck guide latest`;\nthat command can read the archived feature after `current-feature.md` is reset.\n\n## Rules\n\n- The work item is the unit of history: one squashed feature, fix, or rollback\n  commit on main, even if the branch carried several checkpoint commits.\n- A rollback preserves the original feature archive and adds a separate rollback\n  archive. Never rewrite history to make the feature look as if it never existed.\n- Don't merge unfinished or failing work. The documented `Verify` command, or\n  the fallback build and tests, must pass first.\n- Never merge while a P0 or P1 finding is `open` or `fixed` in the ledger. The\n  recorded ways past the gate without code are `accepted` (only by the user's\n  explicit decision, with their reason) or `invalid` (only from re-examination\n  evidence or the user's explicit call); both travel into the archive, never a\n  silent drop.\n- Never merge with a required or explicitly initiated independent review that\n  is missing, pending, changes-requested, malformed, or stale. The user may\n  explicitly cancel a manual review before completion, but the agent never\n  resets or waives it on the user's behalf.\n- Merging and pushing are the user's calls: get an explicit yes for the merge,\n  then ask whether to push main. Do not treat merge approval, `\u002Fcomplete`, or\n  \"looks good\" as permission to push.\n- Push main only after a separate explicit yes to push main in the current chat.\n- One item per completion. If a parent feature still has unchecked sub-features,\n  leave the parent unchecked.\n\n## Formatting\n\nFormat the output to match the project's conventions in\n`blueprint\u002Fcontext\u002Fai-interaction.md`: concise, scannable markdown, with lists for\nenumerations and tables for matrices rather than dense paragraphs.\n",{"name":5,"description":7},"main","0007775402258962018a93be4dbf1cbcbc1536d8",".agents\u002Fskills\u002Fcomplete",3,"https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass","https:\u002F\u002Fgithub.com\u002Fbradtraversy","traversymedia.com","Full stack web developer and online instructor at https:\u002F\u002Ftraversymedia.com and https:\u002F\u002Fstart.dev","2026-10-06 00:26:26",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add bradtraversy\u002Fskillpass --skill complete",{"tool":39,"command":40},"bunx","bunx skills add bradtraversy\u002Fskillpass --skill complete",{"tool":42,"command":43},"pnpm","pnpm dlx skills add bradtraversy\u002Fskillpass --skill complete","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcomplete\u002FSKILL.md",true,[47],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"reference\u002Fcompletion-recovery.md","REFERENCE","参考文档",15932,"https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcomplete\u002Freference\u002Fcompletion-recovery.md",[54,60,66,71,77,83],{"skillNo":55,"name":56,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M782QBBKH3WTBZDE9DP52XH4","plan","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fworkflow-pack\u002Fskills\u002Fplan\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M782Q980Y1KP1J77TNJHD5H8","apply","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fworkflow-pack\u002Fskills\u002Fapply\u002FSKILL.md",[],[],{"skillNo":67,"name":56,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M782Q75AS3P8SW7D819TACVD","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fworkflow-pack\u002F.claude\u002Fskills\u002Fplan\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":6,"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},"01M782Q5WAP9DQH14TEKG5ZFX5","undeclared-network","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fundeclared-network\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M782Q424YF4TAYYX9TCCZCS8","prompt-injection","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fprompt-injection\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M782Q28RQV6MJM98J6EN7XFF","missing-manifest","https:\u002F\u002Fgithub.com\u002Fbradtraversy\u002Fskillpass\u002Fblob\u002Fmain\u002Fpackages\u002Fvalidator\u002Ffixtures\u002Fmissing-manifest\u002FSKILL.md",[],[]]