V0.8.1/agent smith

scripts/validate-skill-templates.ps1

@@ -1164,17 +1164,27 @@ Add-ValidationResult -Results $results -Name 'dotnet-benchmark enforces valid, p
 Add-ValidationResult -Results $results -Name 'Agent Smith protects informational and multi-target EditorConfig remediation' -Action {
     $skill = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/SKILL.md' -GitRef $Ref
     $reference = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/references/dotnet-editorconfig-conformance.md' -GitRef $Ref
+    $skillAuthoring = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/references/skill-authoring.md' -GitRef $Ref
     $evals = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/evals/evals.json' -GitRef $Ref
     $repair = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/scripts/repair-roslyn-multiproject-artifacts.ps1' -GitRef $Ref
     $repairTests = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/agent-smith/scripts/test-repair-roslyn-multiproject-artifacts.ps1' -GitRef $Ref
 
     Assert-Contains -Name 'agent-smith/SKILL.md' -Content $skill -Needle 'every discovery, investigation, retry, and final `dotnet format` command must include both `--severity info` and `--verify-no-changes`'
     Assert-Contains -Name 'agent-smith/SKILL.md' -Content $skill -Needle 'scripts/repair-roslyn-multiproject-artifacts.ps1'
+    Assert-Contains -Name 'agent-smith/SKILL.md' -Content $skill -Needle 'Always analyze the task graph for safe parallelism and concurrency.'
+    Assert-Contains -Name 'agent-smith/SKILL.md' -Content $skill -Needle '**Be concise. Sacrifice grammar for the sake of concision.**'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle 'Batch independent retrieval through one multi-call request where the tool supports it.'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle 'Choose C# and .NET by default for non-trivial reusable scripts'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle '## Required authoring feedback'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle 'Microsoft''s official .NET support policy'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle 'Optimizing skill descriptions'
+    Assert-Contains -Name 'skill-authoring.md' -Content $skillAuthoring -Needle 'Evaluating skill output quality'
     Assert-Contains -Name 'dotnet-editorconfig-conformance.md' -Content $reference -Needle 'dotnet format style "<solution-or-project>"'
     Assert-Contains -Name 'dotnet-editorconfig-conformance.md' -Content $reference -Needle '`dotnet format` defaults to severity `warn`'
     Assert-Contains -Name 'dotnet-editorconfig-conformance.md' -Content $reference -Needle 'Directory application is all-or-nothing at preflight'
     Assert-Contains -Name 'dotnet-editorconfig-conformance.md' -Content $reference -Needle "git grep -n -F 'Unmerged change from project'"
     Assert-Contains -Name 'agent-smith/evals/evals.json' -Content $evals -Needle 'finish fixing all IDE0161 findings in MultiTargeted.sln'
+    Assert-Contains -Name 'agent-smith/evals/evals.json' -Content $evals -Needle 'fetches twelve independent service endpoints sequentially'
     Assert-Contains -Name 'repair-roslyn-multiproject-artifacts.ps1' -Content $repair -Needle 'function Test-LinePrefix'
     Assert-Contains -Name 'repair-roslyn-multiproject-artifacts.ps1' -Content $repair -Needle "pattern = 'whole-document-namespace-conversion'"
     Assert-Contains -Name 'repair-roslyn-multiproject-artifacts.ps1' -Content $repair -Needle "pattern = 'unrecognized'"
@@ -1203,6 +1213,7 @@ Add-ValidationResult -Results $results -Name 'Git visual commits skill enforces
     $commitLanguage = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/git-visual-commits/references/commit-language.md' -GitRef $Ref
     $subjectValidator = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/git-visual-commits/scripts/validate-commit-subject.ps1' -GitRef $Ref
     $subjectTests = Get-FileText -RepoRoot $repoRoot -RelativePath 'skills/git-visual-commits/scripts/test-commit-subject.ps1' -GitRef $Ref
+    $readme = Get-FileText -RepoRoot $repoRoot -RelativePath 'README.md' -GitRef $Ref
 
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'automatic trigger for this skill, not as a casual hint.'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '### Full-Skill Read and Subject Lock'
@@ -1257,6 +1268,12 @@ Add-ValidationResult -Results $results -Name 'Git visual commits skill enforces
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '**Documentation publishing**'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '**Community health/release communication**'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Temporal proximity is not a grouping signal.'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '#### Single-category context quality gate'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Exactly one changed file is the only fast-path exception; skip this gate for that case.'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Did I actually read the whole `git-visual-commits` skill through EOF in this session before classifying this change?'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Then re-check the complete `git status`, `git diff`, and applicable staged diff; enumerate every changed path'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Put a visible line in the commit plan such as `Quality gate:'
+    Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '`yolo` and `auto` do not bypass this gate.'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle '#### Release-adjacent splitting rule'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Concrete example: if one diff updates `Directory.Build.targets`, `Directory.Packages.props`, or `testenvironments.json`,'
     Assert-Contains -Name 'git-visual-commits/SKILL.md' -Content $skill -Needle 'Keep `.nuget/*/PackageReleaseNotes.txt` with the'
@@ -1307,6 +1324,10 @@ Add-ValidationResult -Results $results -Name 'Git visual commits skill enforces
     Assert-Contains -Name 'git-visual-commits/evals/evals.json' -Content $evals -Needle 'Reads SKILL.md completely through EOF before any staging or commit command'
     Assert-Contains -Name 'git-visual-commits/evals/evals.json' -Content $evals -Needle 'Rejects the proposed subject because 📋 is absent from the approved commit-language table'
     Assert-Contains -Name 'git-visual-commits/evals/evals.json' -Content $evals -Needle 'Runs scripts/validate-commit-subject.ps1 before showing the corrected subject and again immediately before passing it to Git'
+    Assert-Contains -Name 'git-visual-commits/evals/evals.json' -Content $evals -Needle 'Triggers the single-category context quality gate because more than one file is being placed in one category'
+    Assert-Contains -Name 'git-visual-commits/evals/evals.json' -Content $evals -Needle 'Recognizes exactly one changed file as the explicit exception and skips the single-category context quality gate'
+    Assert-Contains -Name 'README.md' -Content $readme -Needle '**Single-category context gate**'
+    Assert-Contains -Name 'README.md' -Content $readme -Needle 'Multi-file plans that initially collapse to one category also require a visible full-context quality gate'
 }
 
 Add-ValidationResult -Results $results -Name 'Git visual squash summary skill stays self-contained and shares commit language rules' -Action {

skills/agent-smith/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: agent-smith
 description: >
-  Apply a rigorous, consistent, evidence-driven software-craftsmanship standard across a whole engineering task. Invoke explicitly as `/agent-smith task`, or automatically for design, architecture, implementation, refactoring, review, API compatibility and Semantic Versioning, testing, benchmarking, performance, documentation, security and DevSecOps, CI/CD, delivery, governance, and engineering assessment. For .NET, also use when asked to fix specific IDE or CA diagnostics, clean up EditorConfig findings, make a solution code-style compliant, fix informational IDE diagnostics, or verify `dotnet format` conformance. It performs the requested work, respects repository conventions and user scope, validates before claiming completion, and reports evidence, trade-offs, and risk honestly. Technology-neutral at its core, with optional specialist guidance. Do NOT use for ordinary prose, casual conversation, translation, image generation, or unrelated factual questions.
+  Use this skill to apply a rigorous, evidence-driven software-craftsmanship standard across an engineering task. Invoke explicitly as `/agent-smith task`, or automatically for architecture, implementation, refactoring, review, API compatibility and Semantic Versioning, testing, benchmarking, performance, skill authoring, documentation, security and DevSecOps, CI/CD, delivery, governance, and engineering assessment. For .NET, also use for IDE or CA diagnostic remediation, EditorConfig cleanup, code-style compliance, informational diagnostics, and `dotnet format` conformance. Perform the requested work, respect repository scope and conventions, validate before completion, and report evidence and risk honestly. Technology-neutral core; specialist guidance loads on demand. Do NOT use for ordinary prose, casual conversation, translation, image generation, or unrelated factual questions.
 ---
 
 # Agent Smith
@@ -16,6 +16,12 @@ Apply **one coherent engineering standard** across design, implementation, valid
 
 The standard is technology-neutral. Specialist guidance (including .NET, Git, GitHub, CI/CD, REST, and software-supply-chain security) is loaded only when the task calls for it, and is never imposed on work where it does not apply.
 
+## Critical skill-authoring lock
+
+When creating, modifying, reviewing, or evaluating a skill, read `references/skill-authoring.md` before editing. Inspect the actual skill, applicable repository instructions, real execution traces, repeated work, and failure evidence before recommending changes; if unavailable, make that inspection the first required step. Always analyze the task graph for safe parallelism and concurrency. State how independent retrieval, execution, validation, and grading can use bounded fan-out; encode it in the skill when useful. Keep dependencies, shared mutations, rate-limited calls, and fragile ordered workflows sequential.
+
+For non-trivial reusable scripts and deterministic validators bundled with a skill, choose C# and .NET by default in this .NET-first skill collection. Cross-repository portability alone is not a reason to retain or introduce Python, Bash, or PowerShell. Use another language only for a concrete repository/host constraint, vendor SDK, or materially simpler native operation; state the evidence. Resolve the latest supported .NET LTS dynamically from Microsoft's official support policy when compatible repository SDK/target-framework pins or explicit user constraints do not decide. Do not replace a simpler native command with a C# program merely to satisfy the preference.
+
 ## Critical .NET conformance lock
 
 When the task selects .NET EditorConfig conformance mode, read both `references/dotnet.md` and `references/dotnet-editorconfig-conformance.md` from the activated skill directory before the first formatter command. Do not search for those resources relative to the target repository or improvise the workflow if a required reference cannot be read.
@@ -77,6 +83,7 @@ A task may select **multiple** modes. Load core principles for every invocation,
 | Security & DevSecOps | identity, authorization, secrets, dependencies, pipelines, supply chain, permissions, deployment security | `references/security-and-devsecops.md` |
 | Delivery & repository engineering | CI/CD, Git, branching, repo structure, releases, automation, containers, deployment | `references/delivery-and-repositories.md` |
 | Documentation | public API docs, README, architecture docs, guides, release notes, examples, DocFX | `references/documentation.md` |
+| Skill authoring | creating, modifying, reviewing, describing, or evaluating agent skills and their bundled resources | `references/skill-authoring.md` |
 | Governance | policies, standards, compliance, metrics, enterprise repo governance, guardrails | `references/governance.md` |
 
 **Load .NET guidance only when .NET or C# is actually relevant.** For non-.NET work, apply the core principles and let local conventions govern language-specific detail.
@@ -87,6 +94,7 @@ A task may select **multiple** modes. Load core principles for every invocation,
 - **Benchmark assessment**: core principles + decision framework + performance + implementation + platform reference (e.g. `dotnet.md`) + response contract; agent-handoff template only if delegation is requested.
 - **Public API review**: core principles + decision framework + api-design-and-compatibility + implementation + platform reference + documentation + response contract.
 - **CI/CD pipeline**: core principles + decision framework + security-and-devsecops + delivery-and-repositories + governance (when policy is involved) + response contract.
+- **Skill authoring**: core principles + decision framework (for material choices) + skill authoring + implementation/testing/documentation as applicable. Make concurrency and script-runtime choices explicit.
 - **Scoped .NET diagnostic remediation** (`/agent-smith fix the named IDE and CA diagnostics in src/Codebelt.Core`): core principles + implementation + .NET + .NET EditorConfig conformance + testing. Preserve the user-supplied diagnostic IDs and path through discovery, edits, final verification, build, tests, and reporting.
 
 ## Repository precedence
@@ -122,7 +130,7 @@ For each material finding: Issue → Why it matters → Evidence or reasoning 
 
 ## Response behaviour
 
-Be direct, respectful, and technically defensible. Challenge weak assumptions; preserve good existing decisions; prioritize material issues; avoid empty praise; distinguish recommendation from requirement; avoid exaggerated certainty; explain trade-offs; and avoid unnecessary verbosity for trivial work.
+**Be concise. Sacrifice grammar for the sake of concision.** Prefer clear fragments when they shorten feedback. Remain respectful and technically defensible. Challenge weak assumptions; preserve good existing decisions; prioritize material issues; avoid empty praise; distinguish recommendation from requirement; avoid exaggerated certainty; explain non-obvious trade-offs. Never omit required evidence, validation limits, blockers, compatibility impact, or material risk.
 
 For substantial assessments, use the structure in `references/response-contract.md` (Assessment → Findings → Recommendation → Trade-offs → Validation → Actionable handoff). Do not force that structure onto every response. When producing a formal assessment or a delegation prompt, use `references/engineering-assessment-template.md` or `references/agent-handoff-template.md`.
 
@@ -154,7 +162,7 @@ This skill must not:
 - turn every small task into an architecture exercise;
 - produce advice without completing the requested work when implementation is possible;
 - fabricate evidence or claim unperformed validation;
-- optimize solely for terseness;
+- sacrifice correctness, required evidence, or material context for terseness;
 - broaden the task without justification;
 - introduce dependencies or abstractions without demonstrating value.
 
@@ -174,6 +182,7 @@ Load on demand, per the routing table:
 - `references/security-and-devsecops.md` — identity, secrets, dependencies, pipelines, supply chain.
 - `references/delivery-and-repositories.md` — CI/CD, Git, releases, repository engineering.
 - `references/documentation.md` — documentation as part of the product.
+- `references/skill-authoring.md` — skill design, parallelism, .NET-first scripts, descriptions, and eval loops.
 - `references/governance.md` — policies, standards, and metrics (Intent → Drivers → Metrics → Actions).
 - `references/response-contract.md` — review severity, finding shape, and assessment structure.
 - `references/engineering-assessment-template.md` — fill-in template for a formal assessment.

skills/agent-smith/evals/evals.json

@@ -183,6 +183,22 @@
         "After deliberate recovery, the response requires the artifact tool to report clean and reruns broad dotnet format against MultiTargeted.sln with --severity info and --verify-no-changes",
         "The response does not proceed to build or tests or claim completion while any artifact remains unsafe or detected"
       ]
+    },
+    {
+      "id": 14,
+      "prompt": "/agent-smith improve our incident-research skill in this .NET-first repository family. It fetches twelve independent service endpoints sequentially, includes a reusable Python validator, has a long implementation-focused description, and has never been compared with a baseline. Keep the guidance reusable across our repositories and tell future agents how to validate the change.",
+      "expected_output": "A concise, evidence-driven skill-authoring workflow grounded in real traces and repository constraints. It explicitly maps safe bounded concurrency for independent fetches and eval work, preserves ordered or rate-limited operations, favors a C#/.NET validator using a dynamically resolved supported LTS when local pins do not decide, improves and tests the description as an activation contract, and requires clean-context candidate-versus-baseline evals with objective grading, timing, aggregation, and human review.",
+      "expectations": [
+        "The response requires inspecting the existing skill, repository conventions, real execution traces, and failure evidence before rewriting generic guidance",
+        "The response identifies the twelve independent endpoint calls as concurrency candidates and recommends batching or bounded fan-out instead of sequential fetching",
+        "The response preserves sequential execution for real dependencies, shared mutations, rate limits, or fragile ordering and requires deterministic result ordering and attributable failures",
+        "The response recommends measuring elapsed time, cost or tokens, error rate, and output quality rather than assuming concurrency is faster",
+        "The response favors replacing the reusable Python validator with C# and .NET unless repository or host constraints justify another language",
+        "The response resolves the latest supported .NET LTS dynamically from Microsoft's official support policy when no compatible repository SDK or target-framework pin decides the version, rather than hardcoding a drifting release",
+        "The response treats the description as the activation contract, uses concise imperative user-intent wording, respects the 1,024-character limit, and tests realistic positive and near-miss triggers with repeated runs and a fixed train/validation split",
+        "The response requires realistic clean-context candidate-versus-original baseline evals, objective assertions with deterministic grading where possible, timing capture, benchmark aggregation, and human review",
+        "The feedback is concise and permits compact fragments without omitting validation limits, blockers, compatibility impact, or material risk"
+      ]
     }
   ]
 }