@xevy/skills 0.1.51 → 0.1.52
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.
- package/.claude-plugin/plugin.json +1 -1
- package/README.md +3 -4
- package/bin/cli.mjs +1 -0
- package/package.json +1 -1
- package/skills/create-xevy-task/SKILL.md +2 -2
- package/skills/execute/SKILL.md +7 -7
- package/skills/mark-ready/SKILL.md +2 -2
- package/skills/plan/SKILL.md +23 -34
- package/skills/xevy-setup/SKILL.md +1 -1
- package/skills/analyze/SKILL.md +0 -68
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"name": "xevy",
|
|
3
3
|
"displayName": "Xevy",
|
|
4
4
|
"description": "Native /xevy:* SDLC workflow commands driven through the Xevy MCP server. Analyze, plan, execute, fix, and move-to-qa run locally by default and accept --cloud for the server Web Agent. Merge stays local; merge-pull-request evaluates and lands the remote PR.",
|
|
5
|
-
"version": "0.1.
|
|
5
|
+
"version": "0.1.52",
|
|
6
6
|
"author": {
|
|
7
7
|
"name": "Xevy"
|
|
8
8
|
},
|
package/README.md
CHANGED
|
@@ -112,8 +112,7 @@ with `codex mcp add`.
|
|
|
112
112
|
| `generate-image` | Generate an image and save it in the project Gallery |
|
|
113
113
|
| `seo-research` | Run and explain decision-led SEO reports and supporting research |
|
|
114
114
|
| `xevy-setup` | Link repo, `.xevy/project.json`, deploy defaults, env sync |
|
|
115
|
-
| `
|
|
116
|
-
| `plan [task] [--cloud]` | Planning + review (local by default) |
|
|
115
|
+
| `plan [task] [--cloud]` | Reconnaissance, numbered technical steps, and linked visualization for approval |
|
|
117
116
|
| `execute [task] [--cloud]` | Build, verify, and open a PR (local by default) |
|
|
118
117
|
| `fix [task] [--cloud]` | Rework after review (local by default) |
|
|
119
118
|
| `merge [task]` | Merge a task branch locally for review; leaves remotes and pull requests unchanged |
|
|
@@ -123,7 +122,7 @@ with `codex mcp add`.
|
|
|
123
122
|
| `investigate` | Diagnose with ephemeral scripts (.env; no secrets) |
|
|
124
123
|
| `create-xevy-task` | Capture session work into a task |
|
|
125
124
|
|
|
126
|
-
The
|
|
125
|
+
The plan, execute, fix, and move-to-qa skills run in the current coding-agent session by default. Add the exact `--cloud` flag before or after the task code to delegate one of those workflows to Xevy's server agent, for example `execute VI22-01 --cloud`. Merge stays local; `merge-pull-request` evaluates and lands an open pull request.
|
|
127
126
|
|
|
128
127
|
## Computer and verified guides
|
|
129
128
|
|
|
@@ -137,7 +136,7 @@ Use generate-guide to make docs/guides/customer-form/manual.md with screenshots
|
|
|
137
136
|
|
|
138
137
|
Computer works across desktop and browser apps and projects. Windows uses the signed-in desktop; macOS 14+ requires native Screen Recording and Accessibility consent. The guide workflow stays in the active engine, respects blocks and human take-over, and identifies incomplete verification or publication. Execute/fix use it for UI journeys when an authorized test environment is available.
|
|
139
138
|
|
|
140
|
-
The package version is `0.1.
|
|
139
|
+
The package version is `0.1.52`. Run `npx @xevy/skills update -y` after that version is published.
|
|
141
140
|
|
|
142
141
|
## Security
|
|
143
142
|
|
package/bin/cli.mjs
CHANGED
package/package.json
CHANGED
|
@@ -47,7 +47,7 @@ Check `list_tasks`/`find_task` once more before creating to avoid a duplicate.
|
|
|
47
47
|
After the task exists, if scope, approach, category/epic placement, or trade-offs are **genuinely ambiguous** — don't bury them in the description or guess:
|
|
48
48
|
- Call `mcp__xevy__create_decision` with preset choices + explanations (optional follow-up per choice).
|
|
49
49
|
- If the user is present, use the host's user-question mechanism first, then persist the answer with `create_decision` + `answer_decision`.
|
|
50
|
-
- Open questions gate `/xevy:mark-ready`,
|
|
50
|
+
- Open questions gate `/xevy:mark-ready`, plan and execute until answered — keep them few and high-signal.
|
|
51
51
|
- In **capture mode**, only raise decisions for real ambiguities in requirements or review routing — not for documented AI work.
|
|
52
52
|
- In **ordinary mode**, this is the primary way to capture unresolved intake ambiguities.
|
|
53
53
|
|
|
@@ -62,7 +62,7 @@ After the task exists, if scope, approach, category/epic placement, or trade-off
|
|
|
62
62
|
No AI work to capture (or a fresh, unrelated task): create the task from `$ARGUMENTS`/the request. Capture the user's story, problem, expectation, and desired outcome in plain language — do NOT fabricate technical detail or a PR. Raise any genuine ambiguities with `create_decision` (section 4b). Leave status at `draft` — use `/xevy:mark-ready` to promote once open questions are answered. Do NOT set `ready` while any decision is still open. Skip the review/QA routing.
|
|
63
63
|
|
|
64
64
|
## 7. Report
|
|
65
|
-
Give the task **code + the `url` returned by the create/update tool** (always show this link so the user can open and review it), its **status**, any **open decisions** (call `list_task_decisions` if you raised questions), and a one-line summary of what was captured (PR link, branch, preview URL). Tell the user a developer/QA can now pick it up directly from Xevy (capture mode) or that they should answer open questions and
|
|
65
|
+
Give the task **code + the `url` returned by the create/update tool** (always show this link so the user can open and review it), its **status**, any **open decisions** (call `list_task_decisions` if you raised questions), and a one-line summary of what was captured (PR link, branch, preview URL). Tell the user a developer/QA can now pick it up directly from Xevy (capture mode) or that they should answer open questions and plan while draft (ordinary mode). After planning, AI Agent-owned Process Owner stages advance automatically; human-owned stages wait for the user to run `/xevy:mark-ready`.
|
|
66
66
|
|
|
67
67
|
## Rules
|
|
68
68
|
- Call order: `create_task` first, then any taskId-dependent call with its returned `task.id`; never retry a `Task not found` note call.
|
package/skills/execute/SKILL.md
CHANGED
|
@@ -3,7 +3,7 @@ name: execute
|
|
|
3
3
|
description: Execute code changes locally by default — use the task worktree, implement, push, and open a PR for separate verification and review. Pass --cloud to delegate implementation to the Xevy server agent, then review and correct its result.
|
|
4
4
|
argument-hint: "[task] [--cloud]"
|
|
5
5
|
disable-model-invocation: true
|
|
6
|
-
allowed-tools: "Read Grep Edit Write Bash Agent mcp__xevy__find_task mcp__xevy__update_task mcp__xevy__update_task_status mcp__xevy__list_specs mcp__xevy__create_spec mcp__xevy__update_spec
|
|
6
|
+
allowed-tools: "Read Grep Edit Write Bash Agent mcp__xevy__find_task mcp__xevy__update_task mcp__xevy__update_task_status mcp__xevy__get_task_plan mcp__xevy__list_specs mcp__xevy__create_spec mcp__xevy__update_spec mcp__xevy__list_subtasks mcp__xevy__update_subtask mcp__xevy__list_task_notes mcp__xevy__create_task_note mcp__xevy__update_task_note mcp__xevy__list_sprints mcp__xevy__assign_task_to_sprint mcp__xevy__list_task_decisions mcp__xevy__create_decision mcp__xevy__answer_decision mcp__xevy__start_ai_action mcp__xevy__get_ai_execution mcp__xevy__list_ai_executions set_session_task_lock"
|
|
7
7
|
---
|
|
8
8
|
|
|
9
9
|
# /xevy:execute — execute locally or in the cloud
|
|
@@ -20,20 +20,20 @@ Local implementation is the default. Pass the exact standalone `--cloud` flag to
|
|
|
20
20
|
|
|
21
21
|
## 1. Resolve and apply common gates
|
|
22
22
|
|
|
23
|
-
- Resolve the
|
|
24
|
-
- Sprint gate: if an active sprint exists and the task is not on it, require
|
|
23
|
+
- Resolve the task and read its current numbered Plan and linked visualization. Fetch `get_task_plan` and require approval of the current fingerprint before implementation. Missing, stale, or absent approval means present the short titles for human approval and pause affected implementation. Preserve original requirements and legacy history. An Analysis section and generated subtasks are never prerequisites.
|
|
24
|
+
- Sprint gate: if an active sprint exists and the task is not on it, require the Plan and `storyPoints` greater than zero, then call `assign_task_to_sprint`. Record missing sprint prerequisites as blocked and preserve authorization gates.
|
|
25
25
|
|
|
26
26
|
### Code-only checklist and resume fallback
|
|
27
27
|
|
|
28
|
-
Classify the actual work described in each existing subtask and any resume note before following its next action; keywords and stage labels alone do not determine eligibility. Mark legacy purely investigative, validation, verification, QA, review, or procedural spec-lock items `skipped` with an explanation through `update_subtask`. Preparation belongs to
|
|
28
|
+
Classify the actual work described in each existing subtask and any resume note before following its next action; keywords and stage labels alone do not determine eligibility. Mark legacy purely investigative, validation, verification, QA, review, or procedural spec-lock items `skipped` with an explanation through `update_subtask`. Preparation belongs to Plan; post-execution checks belong to separate **QA Notes** or **Technical Review Instructions**. Select the next pending code subtask when a stale resume target is workflow-only.
|
|
29
29
|
|
|
30
|
-
For mixed items, implement the concrete code portion and defer its workflow checks to those notes. Input-validation implementation and adding/editing test files remain code work. Procedural checklist requirements do not block Execute; existing unanswered
|
|
30
|
+
For mixed items, implement the concrete code portion and defer its workflow checks to those notes. Input-validation implementation and adding/editing test files remain code work. Procedural checklist requirements do not block Execute; existing unanswered Plan decisions, missing authorization, and conflicting technical contracts remain blockers. If inline planning is authorized, finish investigation and feasibility checks before recording code-only implementation steps. If no code changes remain, report that truthfully; never create empty commits or claim skipped work produced implementation.
|
|
31
31
|
|
|
32
32
|
## 2A. Local workflow (default)
|
|
33
33
|
|
|
34
34
|
### Decisions gate
|
|
35
35
|
|
|
36
|
-
Call `mcp__xevy__list_task_decisions` before implementing. Existing unanswered
|
|
36
|
+
Call `mcp__xevy__list_task_decisions` before implementing. Existing unanswered Plan and intake decisions, missing authorization, and scope or contract conflicts remain gates; record affected work as blocked and continue independent authorized work. For new material implementation choices, inspect the code and record `create_decision` with `raisedAtStage: "execute"`, exactly one `recommended: true` choice per question (including follow-ups), and evidence/rationale in explanations. The recommended path is answered by AI; continue without asking the user or re-running Plan.
|
|
37
37
|
|
|
38
38
|
### Worktree protocol
|
|
39
39
|
|
|
@@ -45,7 +45,7 @@ Call `mcp__xevy__list_task_decisions` before implementing. Existing unanswered A
|
|
|
45
45
|
### Implement
|
|
46
46
|
|
|
47
47
|
- Move the task to `in_progress` with `update_task_status` when implementation begins. This mirrors cloud execution, which moves to `in_progress` at start and `for_review` on success.
|
|
48
|
-
- Work through the
|
|
48
|
+
- Work through the approved numbered steps in order, committing coherent code milestones without requiring per-subtask checks or sign-off. When useful and supported, dispatch builder/frontend subagents with the worktree as their working directory. Keep changes surgical, match existing style, and update existing legacy subtask status through MCP only when applicable; never generate subtasks from the Plan.
|
|
49
49
|
- Before editing UI pages, components, app shell, forms, or styling, follow `/xevy:xevy-design` and the project's `./.xevy/design.md`.
|
|
50
50
|
- Apply the Decisions gate policy to material choices discovered during implementation.
|
|
51
51
|
- Record deferred verification, QA, and review in separate handoff notes. A code subtask is complete when its implementation is committed; workflow check completion is outside its checklist.
|
|
@@ -19,11 +19,11 @@ Read the resolved task's status. `mark-ready` only promotes a task that is in **
|
|
|
19
19
|
## 3. Answer open questions (MANDATORY gate)
|
|
20
20
|
- Call `mcp__xevy__list_task_decisions` for the task.
|
|
21
21
|
- For **every** decision that is still OPEN, present the question and its preset choices to the user, then record their answer with `mcp__xevy__answer_decision` (pass the chosen `answerChoiceId` and/or `answerText`). Never fabricate an answer or skip an open question.
|
|
22
|
-
- Answering may trigger a server-side
|
|
22
|
+
- Answering may trigger a server-side Plan refresh — that is expected and does **not** change the task status.
|
|
23
23
|
- Do not proceed while any decision is still open.
|
|
24
24
|
|
|
25
25
|
## 4. Mark ready
|
|
26
26
|
Once **no** open decisions remain, call `mcp__xevy__update_task_status({ taskId, status: "ready" })`.
|
|
27
27
|
|
|
28
28
|
## 5. Report
|
|
29
|
-
Report the task code/title, its new status (`ready`), and the task URL. Don't push the user into `/xevy:plan`
|
|
29
|
+
Report the task code/title, its new status (`ready`), and the task URL. Don't push the user into `/xevy:plan` — let them choose the next step at their own pace.
|
package/skills/plan/SKILL.md
CHANGED
|
@@ -1,12 +1,12 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: plan
|
|
3
|
-
description: Plan a task locally by default —
|
|
3
|
+
description: Plan a task locally by default — inspect the code and write a short numbered technical plan with a linked visualization for human approval. Pass --cloud to delegate planning to the Xevy server agent, then review and correct its result.
|
|
4
4
|
argument-hint: "[task] [--cloud]"
|
|
5
5
|
disable-model-invocation: true
|
|
6
|
-
allowed-tools: "Read Grep Bash Agent mcp__xevy__find_task mcp__xevy__create_spec mcp__xevy__update_spec mcp__xevy__list_specs mcp__xevy__get_spec
|
|
6
|
+
allowed-tools: "Read Grep Bash Agent mcp__xevy__find_task mcp__xevy__create_spec mcp__xevy__update_spec mcp__xevy__list_specs mcp__xevy__get_spec mcp__xevy__create_attachment mcp__xevy__create_attachment_version mcp__xevy__get_task_plan mcp__xevy__approve_task_plan mcp__xevy__update_task mcp__xevy__list_task_notes mcp__xevy__create_task_note mcp__xevy__update_task_note mcp__xevy__list_task_decisions mcp__xevy__create_decision mcp__xevy__answer_decision mcp__xevy__get_process_owners mcp__xevy__update_task_status mcp__xevy__start_ai_action mcp__xevy__get_ai_execution mcp__xevy__list_ai_executions"
|
|
7
7
|
---
|
|
8
8
|
|
|
9
|
-
# /xevy:plan —
|
|
9
|
+
# /xevy:plan — inspect, visualize, and approve
|
|
10
10
|
|
|
11
11
|
Local planning is the default. Pass the exact standalone `--cloud` flag to delegate planning to the Xevy server agent.
|
|
12
12
|
|
|
@@ -18,53 +18,42 @@ Local planning is the default. Pass the exact standalone `--cloud` flag to deleg
|
|
|
18
18
|
- If the Local workflow cannot be completed, stop and report exactly what is blocking it, then let the user decide whether to rerun with `--cloud`. Never switch engines mid-run and never run both.
|
|
19
19
|
- With `--cloud`, follow the **Cloud workflow** and do not duplicate the work locally beyond the review-and-correct step it defines.
|
|
20
20
|
|
|
21
|
-
## 1. Resolve the task
|
|
21
|
+
## 1. Resolve the task and investigate
|
|
22
22
|
|
|
23
|
-
Resolve the
|
|
23
|
+
Resolve the task argument or the in-context task. Read the original requirements, existing Plan, relevant code, similar patterns, and open decisions. Plan owns reconnaissance and feasibility checks; an Analysis section is never a prerequisite. Preserve original requirements and legacy task history.
|
|
24
|
+
|
|
25
|
+
Resolve routine technical choices from the code. Collect meaningful human questions together, persist them with `create_decision` at `raisedAtStage: "plan"`, and record human answers with `answerSource: "human"`. Open decisions block the affected work. Use a scout or independent plan reviewer when supported and authorized.
|
|
24
26
|
|
|
25
27
|
## 2A. Local workflow (default)
|
|
26
28
|
|
|
27
|
-
###
|
|
29
|
+
### Write the concrete Plan
|
|
28
30
|
|
|
29
|
-
|
|
31
|
+
Write the canonical `## Plan` section in the task description. Synchronize an existing spec with kind `plan` when present; separate contract specs are optional when the changes need them. Use approximately five or six numbered steps; each Xevy step follows **Title — concise technical details**. Name the relevant files, functions or components, the change, and resulting behavior in one or two short sentences. Keep the plan proportionate to the task; use fewer steps for small changes.
|
|
30
32
|
|
|
31
|
-
|
|
33
|
+
Example stored in Xevy:
|
|
32
34
|
|
|
33
|
-
|
|
35
|
+
1. Remove Analyze — Remove the action from task controls, runner schemas, prompts, and packaged skills; retain historical executions and decisions.
|
|
36
|
+
2. Simplify Plan — Write numbered technical steps to the task description; remove automatic subtask generation.
|
|
34
37
|
|
|
35
|
-
|
|
36
|
-
- Create ordered subtasks with `mcp__xevy__create_subtask`, one per discrete code change, using fullstack or frontend stages. Group work into reviewable end-to-end pieces; keep frontend and backend work for one capability together.
|
|
37
|
-
- Adding input-validation code and adding/editing test files are code work. Investigation, validation, running verification, QA, review, and procedural spec-lock work are outside the numbered implementation plan and subtask checklist.
|
|
38
|
-
- Create or update separate task notes titled **QA Notes** and **Technical Review Instructions** for post-execution checks. Keep these checks out of subtask descriptions and dependencies.
|
|
38
|
+
Plan creates zero subtasks. Preserve existing subtasks and treat them as legacy guidance. Do not convert numbered steps into subtasks, a trace-matrix document, or a required checklist. Keep verification details in separate **QA Notes** or **Technical Review Instructions** when needed. Estimate `storyPoints` during Plan when the project uses sprint estimates.
|
|
39
39
|
|
|
40
|
-
###
|
|
40
|
+
### Link the visualization
|
|
41
41
|
|
|
42
|
-
|
|
42
|
+
Follow [create-artifact](../create-artifact/SKILL.md) to save a private project Artifact associated with the task UUID. Use self-contained HTML by default. Show the proposed UI, before/after behavior, or a clear flow of the changes; connect it to the numbered plan. Update an existing task visualization through a new attachment version when appropriate. Link the returned stable Artifact URL from the task's `## Plan` section and any existing matching plan spec. Never invent an Artifact URL. If publication is unavailable, report the exact gap and leave the plan awaiting completion.
|
|
43
43
|
|
|
44
|
-
|
|
44
|
+
### Review and request approval
|
|
45
45
|
|
|
46
|
-
|
|
46
|
+
Check every requirement against the inspected code, numbered steps, and visualization; correct gaps before presenting it. Do not create review/sign-off subtasks. Do not start Execute or change task status as a side effect of planning.
|
|
47
47
|
|
|
48
|
-
|
|
48
|
+
Fetch `get_task_plan` and retain the fingerprint of the Plan being presented. In chat show only the short numbered titles, the task link, and visualization link, then request approval. Keep the technical details in Xevy. For the example above, chat shows:
|
|
49
49
|
|
|
50
|
-
|
|
50
|
+
1. Remove Analyze.
|
|
51
|
+
2. Simplify Plan.
|
|
51
52
|
|
|
52
|
-
|
|
53
|
+
After a new affirmative human message approves this exact current plan, fetch `get_task_plan` again and compare its `fingerprint` with the revision presented. Record the retained fingerprint through `approve_task_plan` only when they match. A changed plan requires fresh approval. The approval tool records a human decision; planning agents must never approve their own output. Explain unresolved decisions or missing visualization separately and keep affected work blocked. Draft remains preparation until explicitly submitted to Ready.
|
|
53
54
|
|
|
54
55
|
## 2B. Cloud workflow (`--cloud`)
|
|
55
56
|
|
|
56
|
-
|
|
57
|
-
|
|
58
|
-
- Call `mcp__xevy__start_ai_action({ taskId, action: "plan", stream: true })`; always pass `stream: true`. Report live progress as it arrives. Do not use fire-and-forget or sleep/manual polling loops.
|
|
59
|
-
- Only if the streamed call returns before a terminal status because of a genuine disconnect, resume with `mcp__xevy__get_ai_execution` (use `list_ai_executions` if needed).
|
|
60
|
-
- On failure, surface the output and offer to retry or rerun `/xevy:plan` without `--cloud`.
|
|
61
|
-
|
|
62
|
-
### Review and correct
|
|
63
|
-
|
|
64
|
-
- Read the produced plan spec (`list_specs` → `get_spec`) and subtasks (`list_subtasks`).
|
|
65
|
-
- Build a trace matrix from every task/Analysis requirement to a subtask or plan step. Flag orphaned requirements and unjustified steps. Inspect the code with Read/Grep to confirm referenced files and approaches are real and feasible.
|
|
66
|
-
- Check cross-flow risks such as guards, context rehydration, effects, and async side effects during planning. Put post-execution verification of “must NOT” constraints in separate QA Notes or Technical Review Instructions.
|
|
67
|
-
- Correct the produced plan so numbered steps and subtasks contain only concrete code changes with expected behavior; remove workflow-only plan steps and move their checks to separate handoff notes. Retain input-validation implementation and test-file changes. Fix gaps by editing the spec or adding/reordering code subtasks through MCP. Apply the review passes above: consolidate the complete first-pass findings, then verify corrections and affected flows in the second pass. Record the reason for any further pass; retain unresolved matrix gaps and missing evidence in the verdict.
|
|
68
|
-
- Do not change task status or start execution yourself. The server applies Process Owner automation after a successful run.
|
|
57
|
+
Call `start_ai_action({ taskId, action: "plan", stream: true })` and report live progress. If a genuine disconnect ends the stream early, resume through `get_ai_execution` or `list_ai_executions`. On failure, report the captured output and offer a retry in the selected engine.
|
|
69
58
|
|
|
70
|
-
|
|
59
|
+
Read the canonical task description Plan and any existing plan spec. Inspect referenced code and validate the numbered steps, concise technical details, requirement coverage, real linked task Artifact, and zero generated subtasks. Correct gaps in the task Plan, any existing matching spec, or Artifact; leave legacy subtasks intact. Apply the same short-title chat presentation and explicit human approval procedure as local planning. The server owns status automation and never treats successful planning as approval to Execute.
|
|
@@ -54,4 +54,4 @@ Cursor, and other remote MCP clients use OAuth.
|
|
|
54
54
|
- Read `get_project_workbreakdown`. Only when the project has no categories, create 3–6 user/product-area categories (never technical-layer categories such as Frontend or API) and 1–3 starter epics per category for already-present capabilities. If categories exist, skip this step. Never create tasks.
|
|
55
55
|
|
|
56
56
|
## 7. Report
|
|
57
|
-
Summarise: workspace/project, repo link, deploy build pack + ports, count of env keys added (names only), and the next Xevy skills they can run (`
|
|
57
|
+
Summarise: workspace/project, repo link, deploy build pack + ports, count of env keys added (names only), and the next Xevy skills they can run (`plan`, `execute`, `deploy`; slash-command names may be available in some clients). Do not print secrets.
|
package/skills/analyze/SKILL.md
DELETED
|
@@ -1,68 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
name: analyze
|
|
3
|
-
description: Analyze a task locally in the current coding-agent session by default — inspect the codebase, write the Analysis section and story points, then self-review. Pass --cloud to delegate analysis to the Xevy server agent, then review and correct its result.
|
|
4
|
-
argument-hint: "[task] [--cloud]"
|
|
5
|
-
disable-model-invocation: true
|
|
6
|
-
allowed-tools: "Read Grep Bash mcp__xevy__find_task mcp__xevy__update_task mcp__xevy__list_specs mcp__xevy__list_subtasks mcp__xevy__list_task_notes mcp__xevy__create_task_note mcp__xevy__list_task_decisions mcp__xevy__create_decision mcp__xevy__answer_decision mcp__xevy__start_ai_action mcp__xevy__get_ai_execution mcp__xevy__list_ai_executions"
|
|
7
|
-
---
|
|
8
|
-
|
|
9
|
-
# /xevy:analyze — analyze locally or in the cloud
|
|
10
|
-
|
|
11
|
-
Local analysis is the default. Pass the exact standalone `--cloud` flag to delegate the analysis to the Xevy server agent.
|
|
12
|
-
|
|
13
|
-
## 0. Choose the engine
|
|
14
|
-
|
|
15
|
-
- Parse `$ARGUMENTS` for the exact standalone `--cloud` token and remove that token before resolving the task. Do not treat similar text such as `--cloudy`, `cloud`, or `--cloud=true` as the flag.
|
|
16
|
-
- Without `--cloud`, follow the **Local workflow** and do not call `start_ai_action`. The missing flag is an explicit instruction to stay local: for the whole run, `mcp__xevy__start_ai_action`, `mcp__xevy__get_ai_execution`, and `mcp__xevy__list_ai_executions` are forbidden, and so is any other way of handing the work to the Xevy server agent.
|
|
17
|
-
- Never infer the cloud engine from context. Task status, an earlier cloud run, an open pull request, an existing execution record, an idle local worktree, urgency, or a local step that turns out to be slow or hard are **not** consent. Only the literal flag selects the Cloud workflow.
|
|
18
|
-
- If the Local workflow cannot be completed, stop and report exactly what is blocking it, then let the user decide whether to rerun with `--cloud`. Never switch engines mid-run and never run both.
|
|
19
|
-
- With `--cloud`, follow the **Cloud workflow** and do not duplicate the work locally beyond the review-and-correct step it defines.
|
|
20
|
-
|
|
21
|
-
## 1. Resolve the task
|
|
22
|
-
|
|
23
|
-
Resolve the remaining task argument or the in-context task (`mcp__xevy__find_task` / `find_task`). If ambiguous, ask which task. Read existing specs, subtasks, and notes.
|
|
24
|
-
|
|
25
|
-
## 2A. Local workflow (default)
|
|
26
|
-
|
|
27
|
-
### Decisions gate
|
|
28
|
-
|
|
29
|
-
Call `mcp__xevy__list_task_decisions`. If any decision is **open**, review the questions together using their preset choices and explanations, record each human answer with `mcp__xevy__answer_decision`, and fold the answers into the analysis. Never guess or bypass an open decision.
|
|
30
|
-
|
|
31
|
-
### Inspect the codebase
|
|
32
|
-
|
|
33
|
-
Read the relevant files, including callers, callees, and similar patterns. Identify whether the task involves UI changes; if so, study existing UI patterns and record a `UI/UX Flow` note with `mcp__xevy__create_task_note`. Base decisions on established patterns and ask only about genuine ambiguity.
|
|
34
|
-
|
|
35
|
-
Complete investigation, feasibility checks, and validation of the proposed approach during Analyze/Plan preparation. Record findings and actual unresolved questions here. Later numbered plans and subtasks contain concrete code changes; post-execution verification, QA, and review belong to separate handoff notes and workflow.
|
|
36
|
-
|
|
37
|
-
### Write the Analysis
|
|
38
|
-
|
|
39
|
-
Use `mcp__xevy__update_task` to append a `## Analysis` section without replacing the user's original request:
|
|
40
|
-
|
|
41
|
-
- **Overview:** plain-language summary, before/after behaviour, user impact, and key decisions.
|
|
42
|
-
- **Technical Implementation:** affected files/modules, API/data/UI changes, assumptions, risks, and verification approach.
|
|
43
|
-
- Estimate `storyPoints` (1 trivial · 2 small · 3 medium · 5 large · 8 XL · 13 epic) and set it.
|
|
44
|
-
- Finish inspecting the code and similar patterns, then gather all meaningful human questions together before implementation. Persist each through `mcp__xevy__create_decision` with `raisedAtStage: "analyze"`, preset choices, exactly one `recommended: true` choice per question, evidence/rationale in explanations, and necessary branching follow-ups. Resolve routine choices from the code. If the user is present, collect answers together through the host's user-question mechanism and record each with `answer_decision` and `answerSource: "human"`.
|
|
45
|
-
|
|
46
|
-
### Review and correct
|
|
47
|
-
|
|
48
|
-
Re-read the Analysis against the actual code. Confirm every claim is accurate, affected files exist, and risks and acceptance criteria are concrete. Fix every gap or inaccuracy before finishing. Call `mcp__xevy__list_task_decisions` again and surface any decision that remains open.
|
|
49
|
-
|
|
50
|
-
Report a PASS/PARTIAL verdict, what you corrected, any open decisions, and the task URL. Do not push the user to the next phase.
|
|
51
|
-
|
|
52
|
-
## 2B. Cloud workflow (`--cloud`)
|
|
53
|
-
|
|
54
|
-
### Run the server action
|
|
55
|
-
|
|
56
|
-
- Call `mcp__xevy__start_ai_action({ taskId, action: "analyze", stream: true })`; always pass `stream: true`. The call blocks and streams live per-step progress, so report progress as it arrives. Do not use fire-and-forget or sleep/manual polling loops.
|
|
57
|
-
- Only if the streamed call returns before a terminal status because of a genuine client disconnect, resume with `mcp__xevy__get_ai_execution` until `succeeded | failed | cancelled` (use `list_ai_executions` to find the newest run if needed).
|
|
58
|
-
- If it fails, read the captured output, report the error, and offer to retry or rerun `/xevy:analyze` without `--cloud`.
|
|
59
|
-
|
|
60
|
-
### Review and correct
|
|
61
|
-
|
|
62
|
-
Act as the reviewer of the cloud result:
|
|
63
|
-
|
|
64
|
-
- Re-read the task, specs, subtasks, and the Analysis section the agent wrote.
|
|
65
|
-
- Confirm it covers the plain-language overview, before/after behaviour, user impact, affected files, API/data/UI changes, assumptions, risks, verification, and a `storyPoints` estimate. Inspect referenced code with Read/Grep rather than trusting the write-up.
|
|
66
|
-
- Repair missing, inaccurate, or contradictory material with `mcp__xevy__update_task`; append or repair the Analysis and never delete the user's original request. Set `storyPoints` if absent.
|
|
67
|
-
|
|
68
|
-
Report a PASS/PARTIAL verdict, what you corrected, and the task URL. Do not push the user to the next phase.
|