pair-capability-verify-adoption
Checks code conformity against established project standards. It identifies issues without attempting to fix them.
Install
mkdir -p .claude/skills/pair-capability-verify-adoption && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12081" && unzip -o skill.zip -d .claude/skills/pair-capability-verify-adoption && rm skill.zipInstalls to .claude/skills/pair-capability-verify-adoption
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Checks code and configuration against project adoption files for a given scope. Detection-only: returns conformity/non-conformity list per area without resolving issues. Caller decides resolution. Invocable independently or composed by /pair-process-review and /pair-process-implement.Key capabilities
- →Resolve the scope of the audit (e.g., `tech-stack`, `architecture`, `security`).
- →Check tech stack compliance against `adoption/tech/tech-stack.md`.
- →Verify architecture compliance against `adoption/tech/architecture.md` and ADRs.
- →Audit security compliance using security-related adoption files and guidelines.
- →Check coding standards compliance against `code-design guidelines`.
How it works
The skill reads code and configuration files, compares them against project adoption files for a specified scope, and generates a report on conformity or non-conformity.
Inputs & outputs
When to use pair-capability-verify-adoption
- →Conducting architecture audits
- →Checking security compliance
- →Verifying tech stack adoption
About this skill
/pair-capability-verify-adoption — Adoption Compliance Checker
Check code, configuration, or a PR against the project's adoption files. Detection-only — returns a conformity/non-conformity list per area. Does NOT resolve issues; resolution is the caller's responsibility.
Arguments
| Argument | Required | Description |
|---|---|---|
$scope | Yes | Area to check: tech-stack, architecture, security, coding-standards, infrastructure, all. Determines which adoption files are read. |
Scope-to-Adoption Mapping
| Scope | Adoption File(s) |
|---|---|
tech-stack | adoption/tech/tech-stack.md |
architecture | adoption/tech/architecture.md + adoption/tech/adr/ |
security | Security-related sections in adoption files + security guidelines |
coding-standards | Code design adoption + code-design guidelines |
infrastructure | adoption/tech/infrastructure.md |
all | All of the above |
Algorithm
For each area in scope, follow the check → skip → act → verify pattern.
Step 1: Resolve Scope
- Check: Is
$scopeprovided? - Act: If
$scope = all, expand to all 5 areas:tech-stack,architecture,security,coding-standards,infrastructure. - Act: If single scope, use only that area.
- Verify: Area list determined.
Step 2: Check Tech Stack Compliance
Skip if tech-stack not in resolved scope.
- Check: Does adoption/tech/tech-stack.md exist and contain project-specific content?
- Skip: If adoption file missing or template-only → report
tech-stack: NOT CONFIGUREDand move to next area. - Act: Read the adoption file. Extract adopted technologies with versions. Scan the codebase or PR for:
- Dependencies not listed in tech-stack.md (unlisted imports, package.json entries)
- Version mismatches (code uses a different version than adopted)
- Deprecated or removed technologies still in use
- Verify: Report per finding:
- CONFORMANT: All dependencies match adoption.
- NON-CONFORMANT: List each violation with file, dependency, and reason.
Step 3: Check Architecture Compliance
Skip if architecture not in resolved scope.
- Check: Does adoption/tech/architecture.md exist with project-specific content?
- Skip: If missing → report
architecture: NOT CONFIGUREDand move to next area. - Act: Read architecture adoption and existing ADRs in adoption/tech/adr/. Check code for:
- Pattern violations (e.g., direct DB access bypassing service layer)
- Layer boundary crossings not permitted by the architecture
- Architectural changes without a corresponding ADR
- ADR decisions not reflected in implementation
- Verify: Report CONFORMANT or NON-CONFORMANT with findings.
Step 4: Check Security Compliance
Skip if security not in resolved scope.
- Check: Do security-related adoption files or security guidelines exist?
- Skip: If no security guidelines found → report
security: NOT CONFIGUREDand move to next area. - Act: Check code for:
- Hardcoded secrets or credentials
- Input validation gaps at system boundaries
- Injection vulnerabilities (SQL, command, XSS)
- Improper data handling or logging of sensitive data
- Missing authentication/authorization checks
- Verify: Report CONFORMANT or NON-CONFORMANT with findings.
Step 5: Check Coding Standards Compliance
Skip if coding-standards not in resolved scope.
- Check: Do code-design guidelines or coding standards adoption files exist?
- Skip: If missing → report
coding-standards: NOT CONFIGUREDand move to next area. - Act: Check code for:
- Naming convention violations
- File/module organization deviations from adopted patterns
- Missing or improper error handling patterns
- Test structure violations (1:1 mapping, behavior-based tests)
- Verify: Report CONFORMANT or NON-CONFORMANT with findings.
Step 6: Check Infrastructure Compliance
Skip if infrastructure not in resolved scope.
- Check: Does adoption/tech/infrastructure.md exist with project-specific content?
- Skip: If missing → report
infrastructure: NOT CONFIGUREDand move to next area. - Act: Read infrastructure adoption. Check for:
- CI/CD configuration deviations from adopted pipeline
- Deployment patterns not matching adopted infrastructure
- Environment configuration inconsistencies
- Verify: Report CONFORMANT or NON-CONFORMANT with findings.
Output Format
ADOPTION COMPLIANCE REPORT:
├── Scope: [$scope — expanded areas]
├── Tech Stack: [CONFORMANT | NON-CONFORMANT — N findings | NOT CONFIGURED]
├── Architecture: [CONFORMANT | NON-CONFORMANT — N findings | NOT CONFIGURED]
├── Security: [CONFORMANT | NON-CONFORMANT — N findings | NOT CONFIGURED]
├── Coding Standards:[CONFORMANT | NON-CONFORMANT — N findings | NOT CONFIGURED]
└── Infrastructure: [CONFORMANT | NON-CONFORMANT — N findings | NOT CONFIGURED]
FINDINGS (if any):
1. [area] — [description] ([file:line])
Resolution: [/assess-stack | /record-decision | developer action]
2. ...
RESULT: [ALL CONFORMANT | N non-conformities across N areas]
Resolution Routing
This skill does NOT resolve non-conformities. It routes them to the appropriate resolver:
| Area | Resolver | Rationale |
|---|---|---|
| Tech Stack | /pair-capability-assess-stack | Stack changes go through tech assessment |
| Architecture | /pair-capability-record-decision | Architectural gaps need an ADR |
| Security | Developer action | Security fixes are context-specific |
| Coding Standards | Developer action | Code style fixes are manual |
| Infrastructure | Developer action | Infra changes need manual review |
Composition Interface
When composed by /pair-process-review:
- Input:
/pair-process-reviewinvokes/pair-capability-verify-adoptionwith$scope=allduring the adoption compliance phase. - Output: Returns the compliance report. /pair-process-review incorporates findings into review output.
- NON-CONFORMANT areas contribute to CHANGES-REQUESTED verdict.
- NOT CONFIGURED areas are noted but do not block.
When composed by /pair-process-implement:
- Input:
/pair-process-implementmay invoke/pair-capability-verify-adoptionwith a targeted$scope(e.g.,tech-stack) before commit. - Output: Returns the compliance report. /pair-process-implement resolves non-conformities via /pair-capability-assess-stack or /pair-capability-record-decision.
- Unresolved non-conformities do not HALT — /pair-process-implement warns and continues.
When invoked independently:
- Full interactive flow. Developer receives the compliance report and decides resolution.
- This skill only reads and reports — it does not modify files.
Graceful Degradation
See graceful degradation (adoption file missing → report as NOT CONFIGURED, don't fail; guideline missing → check only what's derivable) for the standard scenarios. Additional cases:
- If
$scope=alland no adoption files exist at all, report: "No adoption files found — run/pair-process-bootstrapto establish project standards." - If the codebase is empty or no code changes to check, report all areas as
CONFORMANT(nothing to violate).
Notes
- This skill is read-only — it inspects code and adoption files but never modifies anything.
- Detection-only contract: the skill identifies non-conformities but never resolves them. Resolution is always delegated to the caller or the appropriate skill.
- Idempotent — see idempotency convention. This skill's check: conformant code is confirmed immediately; non-conformant code returns the same findings until fixed.
- Scope is mandatory to prevent accidental full scans when only a targeted check is needed. Use
allexplicitly for comprehensive checks. - Contextual checking: when invoked on a PR, only changes in the PR are checked, not the entire codebase. When invoked independently, the scope of code analysis is determined by the area.
When not to use it
- →When evaluating new technology choices.
- →When the user expects the skill to resolve non-conformities automatically.
- →When no adoption files or guidelines are configured for a given scope.
Limitations
- →It is detection-only; it does not resolve issues.
- →Requires adoption files and guidelines to be present and configured.
- →The scope argument is mandatory to prevent accidental full scans.
How it compares
This skill provides a structured, automated audit of code and configuration against defined project standards, unlike a manual review process.
Compared to similar skills
pair-capability-verify-adoption side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| pair-capability-verify-adoption (this skill) | 0 | 5mo | No flags | Intermediate |
| golden-principles | 0 | 1mo | Review | Intermediate |
| harness-boundary | 0 | 3mo | No flags | Advanced |
| adversarial-review | 0 | 2mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
golden-principles
rjmurillo
Scan repository for golden principle violations with agent-readable remediation. Enforces GP-001 through GP-008 from .agents/governance/golden-principles.md. Use when auditing compliance, preparing PRs, or running garbage collection scans.
harness-boundary
zkp442910864
Harness 工程边界约束与变更评估。Use when: 评估代码变更影响范围、确认安全边界、检查工程规范合规性、审查变更风险等级。
adversarial-review
dahatake
>
beforemerge-fullstack-architecture-review
adrian-coronel
Code review rules for DRY/SOLID layered architecture in fullstack TypeScript applications. Covers dependency direction, service/repository patterns, factory injection, domain entities, security hardening, performance optimization, and code quality patterns. Use this skill when reviewing, writing, or
agent-v3-security-architect
ruvnet
Agent skill for v3-security-architect - invoke with $agent-v3-security-architect
ai-spec-review
jyjeanne
Review a markdown specification across business logic, architecture, performance, security, testing, DevOps/CI/CD, dependencies, standards, UX, documentation, code quality, and maintainability. Generates a structured review, risk register, test plan, implementation tasks, and dimension scores (0–10)