@gempack/squad-mcp 0.6.5 → 0.8.0

package/shared/Skill-Squad-Dev.md

@@ -1,29 +1,31 @@
 # Skill: Squad Dev
 
 ## Objective
-Development skill that takes a user prompt, builds an implementation plan, runs gated advisory with specialized agents, implements, and consolidates via TechLead-Consolidator. Codex is optional (`/squad --codex`) and may be auto-suggested when the plan is high-risk.
+
+Development skill that takes a user prompt, builds an implementation plan, runs gated advisory with specialized agents, implements, and consolidates via TechLead-Consolidator. Codex is optional (`/squad:implement --codex`) and may be auto-suggested when the plan is high-risk.
 
 ## Skill Name
-`/squad`
+
+`/squad:implement`
 
 ## Agent Registry
 
-| subagent_type              | Purpose                                     |
-| -------------------------- | ------------------------------------------- |
-| `po`                       | Business value, UX, requirements fit        |
-| `tech-lead-planner`        | Pre-implementation trade-offs and viability |
-| `tech-lead-consolidator`   | Post-implementation final verdict           |
-| `senior-architect`         | Boundaries, DI, scalability                 |
-| `senior-dba`               | Queries, migrations, EF, cache              |
-| `senior-developer`         | Correctness, robustness, APIs, observability|
-| `senior-dev-reviewer`      | Readability, idioms, naming                 |
-| `senior-dev-security`      | OWASP, authz, sensitive data                |
-| `senior-qa`                | Test coverage, strategy, reliability        |
+| subagent_type            | Purpose                                      |
+| ------------------------ | -------------------------------------------- |
+| `po`                     | Business value, UX, requirements fit         |
+| `tech-lead-planner`      | Pre-implementation trade-offs and viability  |
+| `tech-lead-consolidator` | Post-implementation final verdict            |
+| `senior-architect`       | Boundaries, DI, scalability                  |
+| `senior-dba`             | Queries, migrations, EF, cache               |
+| `senior-developer`       | Correctness, robustness, APIs, observability |
+| `senior-dev-reviewer`    | Readability, idioms, naming                  |
+| `senior-dev-security`    | OWASP, authz, sensitive data                 |
+| `senior-qa`              | Test coverage, strategy, reliability         |
 
 ## General Flow
 
 ```
-User -> /squad {prompt}
+User -> /squad:implement {prompt}
        |
        v
 [0. Pre-Check]
@@ -86,11 +88,13 @@ Summary + modified files + tests + validations + rollback plan + next steps.
 ## Phase Details
 
 ### Phase 0 — Pre-Check
+
 1. Run `git status`; capture current branch.
 2. If uncommitted, unrelated changes are present, warn and ask the user before proceeding.
 3. Record starting SHA for the delivery report.
 
 ### Phase 1 — Understanding and Risk
+
 - Read `$ARGUMENTS`; detect `--codex`.
 - Classify type: Feature / Bug Fix / Refactor / Performance / Security / Business Rule.
 - Explore the codebase (Glob, Grep, Read) to locate the affected area and patterns.
@@ -188,6 +192,7 @@ Be direct. If the plan is good, say so. Do not invent problems.
 Absorb relevant suggestions and show an adjusted-plan diff summary.
 
 ### Phase 4 — Gate 1: User Approval
+
 Present the final plan and wait for explicit approval. Do not proceed without it.
 
 ### Phase 5 — Advisory Squad
@@ -204,6 +209,7 @@ Present the final plan and wait for explicit approval. Do not proceed without it
 | Business Rule | po, senior-developer, senior-qa                  | +senior-dba if data-bound                                                              |
 
 **Hard conditionals based on touched files:**
+
 - Query / Migration / EF / Cache → `senior-dba` required
 - DI / Boundaries / new project or module → `senior-architect` required
 - Endpoint / Auth / Middleware → `senior-dev-security` required
@@ -237,13 +243,16 @@ Stay inside your ownership. Forward anything outside your scope.
 Send all selected agents in a single message with multiple tool calls so they run in parallel.
 
 ### Phase 6 — Gate 2: Blocker Halt
+
 - Any advisory Blocker → HALT. Surface blockers and ask the user how to proceed (revise plan, accept risk, abort).
 - Major/Minor → proceed, capturing them as acceptance criteria for implementation.
 
 ### Phase 7 — Escalation Round (optional)
+
 If an advisory agent forwarded a Blocker/Major to an agent that was not selected, spawn that missing agent with only that forwarded item (not a full review).
 
 ### Phase 8 — Implementation
+
 - Follow the plan and advisory acceptance criteria.
 - Read → Edit/Write each file. Respect project patterns.
 - Method names in English. No emojis.
@@ -285,6 +294,7 @@ Be direct. If it is good, say so. Do not request cosmetic refactors.
 Spawn `tech-lead-consolidator` with every advisory report and the delivered delta. Consolidator produces the final verdict and rollback plan.
 
 ### Phase 11 — Gate 3: Reject Loop (max 2 iterations)
+
 - APPROVED / CHANGES REQUIRED (non-blocker) → apply fixes, then deliver.
 - REJECTED → apply the fix list, re-run affected agents on the delta, re-consolidate.
 - After 2 iterations, stop and hand the situation to the user.
@@ -338,13 +348,13 @@ Spawn `tech-lead-consolidator` with every advisory report and the delivered delt
 
 ## Skill Parameters
 
-| Parameter | Type   | Default | Description |
-|-----------|--------|---------|-------------|
-| --codex   | flag   | off     | Enable Codex for plan validation and implementation review |
+| Parameter | Type   | Default | Description                                                                        |
+| --------- | ------ | ------- | ---------------------------------------------------------------------------------- |
+| --codex   | flag   | off     | Enable Codex for plan validation and implementation review                         |
 | --quick   | flag   | off     | Quick mode (see below). Trades depth for speed. Mutually exclusive with `--codex`. |
-| squad     | string | auto    | Specific squad or "auto" for detection |
-| plan-only | bool   | false   | Build the plan only, do not execute |
-| verbose   | bool   | false   | Show individual agent reports inline |
+| squad     | string | auto    | Specific squad or "auto" for detection                                             |
+| plan-only | bool   | false   | Build the plan only, do not execute                                                |
+| verbose   | bool   | false   | Show individual agent reports inline                                               |
 
 ## Quick Mode (`--quick`)
 
@@ -352,41 +362,42 @@ Reduced agent set, terse prompts, condensed delivery. Suitable for small fixes,
 
 Phase deltas vs. normal mode:
 
-| Phase | Normal | Quick |
-|-------|--------|-------|
-| Plan | Full plan with risks/decisions/tests | Condensed: objective, files, top 1-2 risks. Skip alternatives table. |
-| Codex plan validation | Opt-in via `--codex` | Force-disabled. `--quick --codex` is rejected. |
-| User approval gate | Always required | Auto-proceed when risk Low AND scope ≤3 files AND no security/data path. Otherwise still gates. |
-| Squad | Auto-detect 3-7 specialists + tech-lead | Hard cap: 1 specialist + tech-lead. Specialist = work-type primary. |
-| Agent prompts | Full template | "Flag only Blocker/Major. ≤200 words. No long template. If clean, reply 'No issues in scope.'" |
-| Codex review | Opt-in | Skipped (force-disabled with `--quick`) |
-| Tech-lead consolidator | Always | Skipped when zero Blocker/Major from specialist |
-| Delivery | Full report | Condensed: objective, files, tests, residual risks |
+| Phase                  | Normal                                  | Quick                                                                                           |
+| ---------------------- | --------------------------------------- | ----------------------------------------------------------------------------------------------- |
+| Plan                   | Full plan with risks/decisions/tests    | Condensed: objective, files, top 1-2 risks. Skip alternatives table.                            |
+| Codex plan validation  | Opt-in via `--codex`                    | Force-disabled. `--quick --codex` is rejected.                                                  |
+| User approval gate     | Always required                         | Auto-proceed when risk Low AND scope ≤3 files AND no security/data path. Otherwise still gates. |
+| Squad                  | Auto-detect 3-7 specialists + tech-lead | Hard cap: 1 specialist + tech-lead. Specialist = work-type primary.                             |
+| Agent prompts          | Full template                           | "Flag only Blocker/Major. ≤200 words. No long template. If clean, reply 'No issues in scope.'"  |
+| Codex review           | Opt-in                                  | Skipped (force-disabled with `--quick`)                                                         |
+| Tech-lead consolidator | Always                                  | Skipped when zero Blocker/Major from specialist                                                 |
+| Delivery               | Full report                             | Condensed: objective, files, tests, residual risks                                              |
 
 Critical-change auto-fallback: if scope touches `auth`, `crypto`, `permissions`, `Program.cs`, `Startup.cs`, migrations, EF mappings, or `appsettings`, fall back to normal mode with warning `Quick mode disabled — change touches security/data layer. Running full workflow.`.
 
 ## Usage Examples
 
 ```
-/squad implement dollar balance endpoint
+/squad:implement implement dollar balance endpoint
 -> No Codex
 -> Plan + Planner -> User approves -> Advisory -> Implement -> Consolidator
 
-/squad --codex fix cache bug in ParameterService
+/squad:implement --codex fix cache bug in ParameterService
 -> With Codex
 -> Plan + Planner -> Codex validates -> User approves -> Advisory -> Implement -> Codex reviews -> Consolidator
 
-/squad refactor ExchangeUsdService to split responsibilities
+/squad:implement refactor ExchangeUsdService to split responsibilities
 -> Plan + Planner -> User approves -> Advisory -> Implement -> Consolidator
 
-/squad --quick rename a misspelled local variable in BalanceController
+/squad:implement --quick rename a misspelled local variable in BalanceController
 -> Quick mode: condensed plan, 1 specialist + tech-lead, terse prompts, condensed delivery
 
-/squad --quick --codex anything
+/squad:implement --quick --codex anything
 -> Error: --quick is mutually exclusive with --codex
 ```
 
 ## Inviolable Rules
+
 1. Every implementation starts from a plan (approved explicitly, or auto-proceeded under `--quick` when risk Low).
 2. Codex only runs with user consent (flag or confirmed auto-suggestion). Never combined with `--quick`.
 3. TechLead-Consolidator always delivers the final verdict in normal mode. Under `--quick`, skipped only when the specialist reports zero Blocker/Major findings.

package/shared/Skill-Squad-Review.md

@@ -1,17 +1,19 @@
 # Skill: Squad Review
 
 ## Objective
+
 Skill that takes a user prompt, interprets intent, selects the relevant agents, distributes tasks in parallel, and consolidates the results through TechLead-Consolidator.
 
 ## Skill Name
-`/squad-review`
+
+`/squad:review`
 
 ## How It Works
 
 ### General Flow
 
 ```
-User -> /squad-review {prompt}
+User -> /squad:review {prompt}
                 |
                 v
       [1. Prompt Analysis]
@@ -56,51 +58,63 @@ User -> /squad-review {prompt}
 The user can request a generic review or target a focus area. The skill maps the intent to the correct squad.
 
 #### Full Squad (default for PR review)
+
 All specialized agents + TechLead-Consolidator.
+
 - **When**: Complete PR review, branch ready to merge
-- **Triggers**: `/squad-review`, `/squad-review PR`, `/squad-review branch`
+- **Triggers**: `/squad:review`, `/squad:review PR`, `/squad:review branch`
 
 #### Code Squad
+
 `senior-dev-reviewer` + `senior-developer` + `senior-qa` + `tech-lead-consolidator`
+
 - **When**: Focused review on code quality and correctness
-- **Triggers**: `/squad-review code`
+- **Triggers**: `/squad:review code`
 
 #### Data Squad
+
 `senior-dba` + `senior-developer` + `tech-lead-consolidator`
+
 - **When**: Changes to queries, migrations, cache, EF
-- **Triggers**: `/squad-review data`
+- **Triggers**: `/squad:review data`
 
 #### Security Squad
+
 `senior-dev-security` + `senior-developer` + `senior-dev-reviewer` + `tech-lead-consolidator`
+
 - **When**: Focused security review
-- **Triggers**: `/squad-review security`
+- **Triggers**: `/squad:review security`
 
 #### Architecture Squad
+
 `senior-architect` + `senior-developer` + `senior-dba` + `tech-lead-consolidator`
+
 - **When**: Structural changes, new modules, large refactors
-- **Triggers**: `/squad-review arch`
+- **Triggers**: `/squad:review arch`
 
 #### Business Squad
+
 `po` + `senior-developer` + `senior-qa` + `tech-lead-consolidator`
+
 - **When**: New feature, business-rule change
-- **Triggers**: `/squad-review business`
+- **Triggers**: `/squad:review business`
 
 ### Automatic Squad Detection
 
 When the user does not specify a squad, the skill analyzes the modified files to infer it:
 
-| Modified Files | Selected Agents |
-|----------------|-----------------|
-| Controllers, DTOs, Requests, Responses | senior-developer, senior-dev-security, po |
-| Services (business logic) | senior-developer, senior-dev-reviewer, po, senior-qa |
-| Repositories, Queries | senior-dba, senior-developer |
-| Migrations, Schema | senior-dba, senior-architect |
-| Startup, Program.cs, DI | senior-architect, senior-dev-security |
-| appsettings, configs | senior-dev-security, senior-architect |
-| Tests | senior-qa, senior-dev-reviewer |
-| Middlewares, Filters | senior-dev-security, senior-architect, senior-developer |
-| Dockerfile, pipeline, CI/CD | tech-lead-consolidator |
-| Multiple layers | Full Squad |
+| Modified Files                         | Selected Agents                                         |
+| -------------------------------------- | ------------------------------------------------------- |
+| Controllers, DTOs, Requests, Responses | senior-developer, senior-dev-security, po               |
+| Services (business logic)              | senior-developer, senior-dev-reviewer, po, senior-qa    |
+| Repositories, Queries                  | senior-dba, senior-developer                            |
+| Migrations, Schema                     | senior-dba, senior-architect                            |
+| Startup, Program.cs, DI                | senior-architect, senior-dev-security                   |
+| appsettings, configs                   | senior-dev-security, senior-architect                   |
+| Tests                                  | senior-qa, senior-dev-reviewer                          |
+| Middlewares, Filters                   | senior-dev-security, senior-architect, senior-developer |
+| Dockerfile, pipeline, CI/CD            | tech-lead-consolidator                                  |
+| Multiple layers                        | Full Squad                                              |
 
 TechLead-Consolidator is mandatory in any squad.
 
@@ -156,6 +170,7 @@ If an agent did not participate, mark as "Not evaluated".
 ## Implementation
 
 ### Step 1: Collect Context
+
 ```
 - git status
 - git diff master...HEAD (or specified branch)
@@ -164,6 +179,7 @@ If an agent did not participate, mark as "Not evaluated".
 ```
 
 ### Step 2: Determine Squad
+
 ```
 - If the user specified one, use it
 - Otherwise, analyze modified files and infer the squad
@@ -171,6 +187,7 @@ If an agent did not participate, mark as "Not evaluated".
 ```
 
 ### Step 3: Spawn Agents in Parallel
+
 ```
 - Use the Agent tool with subagent_type for each agent
 - All in parallel (single message with multiple tool calls)
@@ -178,18 +195,21 @@ If an agent did not participate, mark as "Not evaluated".
 ```
 
 ### Step 4: Collect Results
+
 ```
 - Wait for all agents to finish
 - Gather every report
 ```
 
 ### Step 5: Consolidate via TechLead-Consolidator
+
 ```
 - Spawn tech-lead-consolidator with every report
 - Consolidator produces the final verdict
 ```
 
 ### Step 6: Present to the User
+
 ```
 - Display the consolidated report
 - If any agent raised a Blocker, highlight it at the top
@@ -226,12 +246,12 @@ If an agent did not participate, mark as "Not evaluated".
 
 ## Skill Parameters
 
-| Parameter | Type   | Default | Description |
-|-----------|--------|---------|-------------|
-| squad     | string | auto    | Specific squad or "auto" for automatic detection |
-| scope     | string | branch  | "branch" (branch diff), "file:path" (specific file), "commit:hash" |
-| base      | string | master  | Base branch for the diff |
-| verbose   | bool   | false   | If true, show full individual reports; if false, only the consolidated one |
+| Parameter | Type   | Default | Description                                                                        |
+| --------- | ------ | ------- | ---------------------------------------------------------------------------------- |
+| squad     | string | auto    | Specific squad or "auto" for automatic detection                                   |
+| scope     | string | branch  | "branch" (branch diff), "file:path" (specific file), "commit:hash"                 |
+| base      | string | master  | Base branch for the diff                                                           |
+| verbose   | bool   | false   | If true, show full individual reports; if false, only the consolidated one         |
 | --quick   | flag   | off     | Quick mode (see below). Trades depth for speed. Mutually exclusive with `--codex`. |
 
 ## Quick Mode (`--quick`)
@@ -240,14 +260,14 @@ Reduced agent set, terse prompts, condensed output. Goal: usable verdict in roug
 
 Phase deltas vs. normal mode:
 
-| Aspect | Normal | Quick |
-|--------|--------|-------|
-| Agents | Auto-detect 3-7 specialists + tech-lead | Hard cap: 1 specialist + tech-lead. Specialist defaults to `senior-dev-reviewer` (or focus mode primary) |
-| Per-agent prompt | Full template | "Flag only Blocker/Major in your domain. ≤200 words. No scorecard. No comments-by-file table. If clean: 'No issues in scope.'" |
-| Tech-lead consolidator | Always runs | Skipped when zero Blocker/Major reported by specialist |
-| Codex | Opt-in via `--codex` | Force-disabled. `--quick --codex` rejected. |
-| Critical-change auto-fallback | N/A | Diff touching `auth`, `crypto`, `permissions`, `Program.cs`, `Startup.cs`, migrations, `appsettings` falls back to normal mode with warning |
-| Output | Full Markdown with scorecard, per-file comments, individual reports | Condensed: verdict + top 3 issues + "run without --quick for full review" hint |
+| Aspect                        | Normal                                                              | Quick                                                                                                                                       |
+| ----------------------------- | ------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
+| Agents                        | Auto-detect 3-7 specialists + tech-lead                             | Hard cap: 1 specialist + tech-lead. Specialist defaults to `senior-dev-reviewer` (or focus mode primary)                                    |
+| Per-agent prompt              | Full template                                                       | "Flag only Blocker/Major in your domain. ≤200 words. No scorecard. No comments-by-file table. If clean: 'No issues in scope.'"              |
+| Tech-lead consolidator        | Always runs                                                         | Skipped when zero Blocker/Major reported by specialist                                                                                      |
+| Codex                         | Opt-in via `--codex`                                                | Force-disabled. `--quick --codex` rejected.                                                                                                 |
+| Critical-change auto-fallback | N/A                                                                 | Diff touching `auth`, `crypto`, `permissions`, `Program.cs`, `Startup.cs`, migrations, `appsettings` falls back to normal mode with warning |
+| Output                        | Full Markdown with scorecard, per-file comments, individual reports | Condensed: verdict + top 3 issues + "run without --quick for full review" hint                                                              |
 
 Quick agent prompt:
 
@@ -297,41 +317,44 @@ Verdict: APPROVED — no Blocker or Major findings.
 ## Usage Examples
 
 ```
-/squad-review
+/squad:review
 -> Auto-detects the squad; reviews diff of current branch vs. master
 
-/squad-review security
+/squad:review security
 -> Security squad; reviews current branch diff
 
-/squad-review arch Services/ParameterService.cs
+/squad:review arch Services/ParameterService.cs
 -> Architecture squad; focused on the specified file
 
-/squad-review full
+/squad:review full
 -> Every agent; complete review
 
-/squad-review --quick
+/squad:review --quick
 -> Quick mode: 1 specialist + tech-lead, terse prompts, condensed output
 
-/squad-review --quick code
+/squad:review --quick code
 -> Quick code-quality review (senior-dev-reviewer + tech-lead)
 
-/squad-review --quick --codex
+/squad:review --quick --codex
 -> Error: --quick is mutually exclusive with --codex
 ```
 
 ## Considerations
 
 ### Performance
+
 - Parallel agents minimize total time
 - TechLead-Consolidator runs sequentially after the others (needs their reports)
 - For large reviews, consider splitting by area or module
 
 ### Cost
+
 - Each agent consumes tokens independently
 - Full squad = 7 specialist agents + TechLead-Consolidator = 8 calls
 - For small changes, auto-selection avoids unnecessary agents
 
 ### Limitations
+
 - Agents do not talk to each other (only through TechLead-Consolidator)
 - Forwarded items are informational — they do not trigger new executions in this skill
 - TechLead-Consolidator consolidates but does not re-execute agents

package/shared/_Severity-and-Ownership.md

@@ -41,12 +41,12 @@ Shared reference for the squad. Each topic has a single primary owner. Other age
 
 Every finding carries a severity. Agents must use these labels consistently.
 
-| Level      | Meaning                                                                                    | Merge Impact                 |
-| ---------- | ------------------------------------------------------------------------------------------ | ---------------------------- |
-| Blocker    | Cannot ship: correctness break, security hole, data loss, production outage likely         | Halts merge                  |
-| Major      | Significant risk or violation with no reasonable justification                             | Halts merge unless justified |
-| Minor      | Quality issue, local smell, limited impact                                                 | Does not block               |
-| Suggestion | Improvement opportunity, nitpick, or stylistic preference                                  | Does not block               |
+| Level      | Meaning                                                                            | Merge Impact                 |
+| ---------- | ---------------------------------------------------------------------------------- | ---------------------------- |
+| Blocker    | Cannot ship: correctness break, security hole, data loss, production outage likely | Halts merge                  |
+| Major      | Significant risk or violation with no reasonable justification                     | Halts merge unless justified |
+| Minor      | Quality issue, local smell, limited impact                                         | Does not block               |
+| Suggestion | Improvement opportunity, nitpick, or stylistic preference                          | Does not block               |
 
 ## Standard Section: Assumptions and Limitations