@xevy/skills 0.1.69 → 0.1.71
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"name": "xevy",
|
|
3
3
|
"displayName": "Xevy",
|
|
4
4
|
"description": "Native Xevy SDLC and Features workflows through MCP: plan, implement, review, record QA evidence and publish verified screenshot guides.",
|
|
5
|
-
"version": "0.1.
|
|
5
|
+
"version": "0.1.71",
|
|
6
6
|
"author": {
|
|
7
7
|
"name": "Xevy"
|
|
8
8
|
},
|
package/package.json
CHANGED
package/skills/execute/SKILL.md
CHANGED
|
@@ -39,6 +39,8 @@ Call `mcp__xevy__list_task_decisions` before implementing. Existing unanswered P
|
|
|
39
39
|
|
|
40
40
|
### Worktree protocol
|
|
41
41
|
|
|
42
|
+
- In a task Dev Session, reuse the task branch and worktree already registered at startup. Register an additional task through `register_dev_session_branch` only when needed, and edit its returned directory after verified success. Retry a failed registration using the same `branchId`; listing branches or waiting does not complete registration.
|
|
43
|
+
|
|
42
44
|
- Record the current branch and directory. Derive a canonical branch `xevy/<TASK-ID>-<slug>` using the task code or the first eight characters of its UUID. Reuse an existing task branch; never create a second branch for the same task.
|
|
43
45
|
- Create or attach a git worktree in a sibling directory and change into it. When resuming, first merge the latest source branch (for example `git merge origin/main`). Do all work in the task worktree.
|
|
44
46
|
- Verify the absolute worktree path with `git rev-parse --show-toplevel` and its branch with `git branch --show-current` before editing. Keep the task code, UUID, title, absolute worktree path, and branch in the session context. On resume or after changing directories, verify and refresh that context before continuing.
|
|
@@ -9,15 +9,15 @@ Review the exact PR and report evidence-based findings. Review remains **read-on
|
|
|
9
9
|
|
|
10
10
|
## 1. Identify the PR and working directory
|
|
11
11
|
|
|
12
|
-
- **Triage review depth first.** Inventory all changed files and affected flows, then classify each as focused (small, isolated, low-impact), standard (ordinary behavior change), or deep (security/auth, data integrity, payments/business decisions, concurrency/retries, migrations, deployment, broad refactors, or other critical paths). This adjusts effort, not coverage: account for every changed file and material flow; uncertain or high-impact areas receive deep review. Keep small PRs focused and avoid redundant historical or test work that cannot affect the verdict.
|
|
12
|
+
- **Triage review depth first.** Inventory all changed files and affected flows, then classify each as focused (small, isolated, low-impact), standard (ordinary behavior change), or deep (security/auth, data integrity, payments/business decisions, concurrency/retries, migrations, deployment, broad refactors, or other critical paths). This adjusts effort, not coverage: account for every changed file and material flow; uncertain or high-impact areas receive deep review. Keep small PRs focused and avoid redundant historical or test work that cannot affect the verdict. Use the host's standard reasoning tier for focused and standard reviews; reserve the highest tier for PRs with a deep area.
|
|
13
13
|
- **Keep a compact internal coverage ledger.** Track each changed file/flow, risk tier, reviewer/pass, evidence inspected or tested, and remaining gaps. Account for every changed file and requirement before the verdict. If material gaps mean the full changed scope cannot be reviewed, the verdict must be **MERGE SAFETY UNVERIFIED**. Keep the full ledger internal unless requested.
|
|
14
14
|
- **Parallelize when useful and supported.** For substantial PRs with separable behavior/risk areas and available independent reviewers, delegate disjoint scopes after pinning identical head/base SHAs and sharing task intent and conventions. Run independent safe checks concurrently where supported. Keep small PRs with one reviewer when coordination costs outweigh benefit. The lead owns the full inventory, confirms all delegated results arrived, reconciles disagreements against source evidence, and issues the verdict. Delegation never excuses skipped scope.
|
|
15
|
-
- **Add an omission-focused independent pass for
|
|
15
|
+
- **Add an omission-focused independent pass only for PRs with a deep area, or when the user asks.** Give the second reviewer the pinned diff, requirements, changed-flow inventory, and ledger; ask them to find omissions, not merely confirm existing findings. Probe applicable error/recovery paths, authorization, concurrency/retries, partial data, and cross-component contracts. Focused and standard PRs use one reviewer, and the lead closes with a short explicit gap pass; the lead does the same when a second reviewer is unavailable and does not claim independent review.
|
|
16
16
|
- Accept a PR URL, number, or branch. If omitted, use the current branch's PR only when exactly one open PR matches; otherwise ask which PR.
|
|
17
17
|
- Identify the repository from the PR URL/metadata; do not ask the user to select a project directory or worktree when the PR uniquely identifies them. Ask only if repository identity is ambiguous or inaccessible.
|
|
18
18
|
- Review the PR's remote source and target branches, pinning their actual SHAs and merge-base. For required local verification, automatically locate an existing Git worktree where the PR source branch is checked out. Use that worktree only if it belongs to this PR/source branch and is safe to use; do not disrupt unrelated changes or another active task. If no suitable worktree exists, create a dedicated worktree for the PR source branch and record its path. Never ask the user which worktree to use. Do not switch branches in an existing checkout.
|
|
19
19
|
- Read the applicable `AGENTS.md` / `AGENTS.override.md` chain from the selected worktree. Keep inspection non-mutating: do not edit tracked files, commit, push, or merge. Only the final summary-comment step below may post a comment. Fetch/update remote refs as needed without altering another checkout. If the PR source branch is not available locally, fetch it and create the worktree at the exact reviewed head SHA; ensure the worktree represents the PR head, not a stale local branch.
|
|
20
|
-
- Testing is required. In the PR source worktree, run the
|
|
20
|
+
- Testing is required. In the PR source worktree, run the checks relevant to the changed areas when feasible (selected below); include regression tests for suspected issues where practical. Avoid exposing secrets and do not run destructive commands. If dependencies, credentials, environment, or infrastructure prevent a check, report the exact limitation and mark that verification unverified; do not imply tests passed or issue an unqualified safe-to-merge verdict. Preserve pre-existing worktree changes and do not clean/reset them.
|
|
21
21
|
- Check whether this session is locked to a different task; do not switch its scope without the user's explicit approval under the session-lock rules.
|
|
22
22
|
- For a remote Dev Session, use `read_dev_session_pull_request` with the linked repository ID and PR number to read authenticated PR evidence; select the `dev_session` capability group if needed. Outside a session, use `code_pull_request` with the project and linked repository. Read `metadata`, then paginate `files`, `commits`, `reviews`, and `checks` until `nextPage` is null; inspect `diff` and every `truncated` flag. Use the returned top-level `headSha` and `baseSha` for current commit identity; `data.base.sha` retains GitHub's recorded PR base and can differ from the current target. Treat truncated or unavailable evidence as incomplete. Credentials stay in the toolbox; these tools require no `gh` login or token in the agent shell. If access is expired, ask the GitHub connection owner to reconnect and retry.
|
|
23
23
|
- For local verification in a remote Dev Session, use `prepare_dev_session_pull_request_review` with that repository ID and PR number. It fetches the exact PR head and current target into a fresh detached checkout and returns `dir`, `headSha`, `baseSha`, and `mergeBaseSha`. Run inspection and tests in `/workspace/<dir>`; preserve the original session checkout. The PR target comes from GitHub metadata and can differ from the session's configured base. Follow the returned evidence and report any incomplete fetch or history.
|
|
@@ -42,10 +42,24 @@ Review the exact PR and report evidence-based findings. Review remains **read-on
|
|
|
42
42
|
- Investigate each suspected issue before reporting it: first establish intended behavior from the linked task and acceptance criteria, explicit decisions, PR description, design/spec documentation, and established surrounding behavior; then trace reachable callers, affected users/data, triggering conditions, existing guards, and consequences if left unfixed. Use a safe reproduction or test where feasible. Do not call an intentional change or omission a defect merely because a control, feature, behavior, or test is absent compared with the base branch. For example, a removed button is not a finding if its removal is required or explicitly intended. Report only behavior that conflicts with established intent or is an unintended consequence of the change as a confirmed finding. If intent is ambiguous or cannot be resolved, put it under **Warnings / Questions** with the missing evidence and potential impact; do not state it as a defect or release blocker. Classify each confirmed issue as **Release blocker** or **Non-blocking** independently of severity. A release blocker prevents a required flow, violates a critical task requirement, causes materially incorrect decisions, or creates significant security, data-integrity, or operational risk. Explain the concrete unfixed consequence; do not infer blocker status from severity alone. If reachability or impact cannot be established, report it under Questions/Risks with blocker status unverified, not as a confirmed blocker.
|
|
43
43
|
- Identify risky changes even when no defect is confirmed: destructive/bulk data operations; auth and permissions; financial or business-decision logic; API/integration contracts; jobs, retries and concurrency; deployment/configuration/dependencies; critical UI actions; broad refactors or unrelated/inherited scope. Assess affected users/data, reversibility, compatibility, test coverage, and recovery safeguards. Do not treat a risk category alone as a defect or release blocker; escalate only when evidence supports an unsafe condition, and mark material unresolved safety questions unverified.
|
|
44
44
|
- Build a risk-based test matrix from every changed behavior: expected outcome, counterexample, overlapping/multiple records, ordering, and relevant boundary conditions. Identify what was exercised by local tests or CI, what was assessed only by code inspection, and what remains unverified. Prioritize tests for high-impact paths over indiscriminate test runs. A passing test suite does not replace tracing untested combinations.
|
|
45
|
-
- On a repeat review, obtain the previous review report (ask the user for it if not accessible) and record its reviewed head/base SHAs, findings, coverage ledger, and unverified items. Recheck every previous finding
|
|
46
|
-
- Run relevant tests,
|
|
45
|
+
- On a repeat review, obtain the previous review report (ask the user for it if not accessible) and record its reviewed head/base SHAs, findings, coverage ledger, and unverified items. Recheck every previous finding, then review the diff from the previously reviewed head to the new head plus every flow that diff touches. Repeat the full changed-file and affected-flow inventory only when the base moved materially, the requirements changed, or the new diff touches shared contracts/interfaces. Carry prior coverage forward for untouched files/flows and record it in the coverage ledger; reopen any area whose dependencies or assumptions changed, and mark unavailable or stale coverage unverified. Never infer safety solely from unchanged lines. Attribute each new finding as introduced by later changes, exposed by a base change, pre-existing but missed, or unknown; do not claim regression origin without evidence.
|
|
46
|
+
- Run only the checks relevant to the changed areas — targeted tests, type-check of changed packages, lint of changed files — where safe in the permitted working directory; avoid commands that rewrite tracked files or require secrets. Run a full build or full suite only for PRs with a deep area or when CI has not run on the reviewed head. Select checks from the risk-based matrix and run independent checks concurrently when safe and supported. If not run, explain why. Separate PR-introduced failures from baseline/infrastructure failures where evidence supports it. Never claim there are no code issues solely because CI is green.
|
|
47
47
|
- Immediately before the verdict, re-read the PR's head and base SHAs. If either changed during review, refresh affected diff, mergeability, checks, and verification or mark the verdict stale/inconclusive; never present an earlier SHA's results as current.
|
|
48
48
|
|
|
49
|
+
### Migration-history safety (when triggered)
|
|
50
|
+
|
|
51
|
+
- **Trigger on behavior, not filenames.** Apply this check when the PR changes migrations, schema, journals or equivalent history metadata, migration configuration, runner code, or startup/deployment migration execution. Include relevant unchanged runner and guard code in the review; a schema or deployment change can alter upgrade safety without adding a migration file.
|
|
52
|
+
- **Establish the actual runner contract from evidence.** Trace the configured entrypoint, resolved runner version/implementation, and startup/deployment wrappers. Identify the authoritative repository journal or equivalent, the database applied-history table/ledger, and the exact eligibility and ordering rules. Distinguish individual migration identity tracking from a latest-timestamp/version watermark; determine strict versus inclusive comparisons, tie handling, checksums/content hashes (including whether they are enforced), dependencies, and reconciliation/startup guards. Cite the relevant code/configuration and source/current-target SHAs; do not impose one framework's rules on another. Filenames and numbering alone never establish eligibility or a defect.
|
|
53
|
+
- **Compare source and current-target histories.** Compare authoritative metadata at the pinned source and current target, using the merge-base to attribute changes. Check lower and equal timestamps/versions, duplicate identities or timestamps, content-hash collisions/mismatches, missing journal entries or migration bodies, dependency gaps/order, and changed or removed already-published identities/content. Inspect relevant concurrently open PRs for competing identities/order when accessible; state that coverage limit when unavailable. Do not infer a conflict solely from similar filenames, or assume that a journal records a checksum the runner actually validates.
|
|
54
|
+
- **Separate branch history from applied database state.** A migration present in the target branch is not proof it was applied to any database. Label applied-state evidence with its environment, provenance and observation time when available; otherwise describe it as a hypothetical upgrade precondition, not a deployment fact. Never infer deployment state from merge order, CI success, or repository metadata alone.
|
|
55
|
+
- **Assess the late-arriving upgrade path.** Evaluate a database where a later migration has already been applied before an incoming earlier migration arrives, including incoming timestamps equal to the applied watermark. Use the runner's actual comparison, identity, dependency and guard rules to determine eligibility and consequences. A clean Git merge or successful fresh installation is not sufficient proof of upgrade safety. Use isolated fixtures/tests (with explicitly disposable state) or a reproducible code trace of the resolved runner and wrappers; never run a migration against a shared, staging, or production database. If isolation cannot be established, use code evidence and state the missing execution proof. Do not claim an isolated test was run when only tracing code.
|
|
56
|
+
- **Distinguish skip, SQL failure, and fail-closed refusal.** Neutral example: an incoming migration has `t_incoming <= t_applied`, while a later migration is already recorded as applied. These are conditional outcomes to investigate, not universal runner behavior:
|
|
57
|
+
- **Silent skipping:** if eligibility requires a timestamp strictly above the watermark, the incoming migration can remain unapplied while the runner reports success. Trace the resulting missing schema/data and affected application flow. **Error if not fixed: No** applies only when evidence establishes no error in that investigated flow; a later failed request is **Yes**, not a harmless skip.
|
|
58
|
+
- **SQL failure:** an individual-identity runner may still attempt the incoming migration, but a missing dependency or incompatible already-upgraded schema can make its SQL fail. Establish the reachable statement/error, transaction or partial-application consequences, and effect on startup/deployment; individual tracking alone does not prove safety. Use **Error if not fixed: Yes** only with demonstrated or code-established failure, otherwise **Unverified**.
|
|
59
|
+
- **Fail-closed refusal:** a reconciliation/startup guard may reject the history before SQL runs. Explain the guard condition, refusal/error, and resulting startup/deployment availability impact. Distinguish intentional protective behavior from a guard defect: a proven refusal of the required upgrade path is a release blocker even when the guard correctly prevents corruption. Never recommend bypassing it. Use **Error if not fixed: Yes** for an established refusal; otherwise **Unverified**.
|
|
60
|
+
- **Apply evidence-based verdicts.** Confirmed unsafe upgrade behavior (silent omission, SQL failure, or refusal preventing the required startup/deployment) is a **Release blocker** and requires **NOT SAFE TO MERGE**; cite the trigger, evidence and concrete unfixed consequence. Missing material runner/history/upgrade evidence requires **MERGE SAFETY UNVERIFIED** unless another established blocker or procedural gate already requires **NOT SAFE TO MERGE**. Put uncertain outcomes in Questions/Risks, not confirmed findings. Numbering differences alone are not a finding. Report which conclusions came from code versus isolated tests and what applied-state evidence remains unverified; a fully verified safe individual-identity upgrade is not a defect merely because its numbering differs.
|
|
61
|
+
- **No live repair or extra mutations.** The review remains read-only for code, records, and merge, with exactly one final PR summary comment by default (unless the user opts out). Do not auto-repair migration history, rewrite journals or applied ledgers, renumber applied migrations, change published identities/checksums, mark migrations applied, or bypass reconciliation/startup guards. Do not migrate shared databases, merge, deploy, or publish packages during review. Describe the smallest safe corrective approach for separately authorized implementation; do not execute it. Fold all migration findings and limitations into the single completed verdict/comment, never a separate mid-review comment.
|
|
62
|
+
|
|
49
63
|
## 4. Deliver a verdict
|
|
50
64
|
|
|
51
65
|
- Use plain-text headings and numbered or bulleted lists only; **never use Markdown tables**. Keep the developer-facing response short, actionable, and easy to copy and paste. Do the full review and maintain internal ledgers, but do not print exhaustive inventories or checklists unless requested. Avoid repeating evidence across sections.
|