advance-technologies-foundation/creatioclient/agentic-code-review
Claude Code and agent skill by advance-technologies-foundation/creatioclient
- community
Checking that the SKILL.md file is still in the repository.
Run a comprehensive multi-perspective review of CreatioClient branch, pull-request, commit, or architecture changes. Use for pre-PR, pre-merge, branch, PR, comprehensive, parallel, or agentic review requests; do not use for ordinary implementation unless a repository review gate is due.
Install
npx vskill@latest install advance-technologies-foundation/creatioclient/agentic-code-reviewSKILL.md
Agentic Code Review
Review the requested change against its stated outcome and CreatioClient's compatibility, authentication, and transport invariants. Findings are the deliverable; do not modify the candidate change.
Workflow
- Read
AGENTS.mdfrom the repository root. - Restate one shared intent brief containing the requested outcome, explicit constraints, and proof boundary.
- Resolve the review scope:
- branch or pre-PR: diff the merge base with
origin/mainagainstHEAD; - pull request: verify the live base and head, then use the exact head diff;
- incremental: review only the named commit or newly pushed commit range;
- architecture: review the proposal plus the smallest relevant source context.
- Record the base/head SHAs, commits, diff stat, and changed files. Include uncommitted and staged changes only
when the user puts them in scope.
- From the repository root, read
.ai/skills/agentic-code-review/references/reviewer-roles.md. - Run independent reviewers in parallel when subagents are available and authorized. Always cover code
quality, security, performance, testing, bugs, intent, and KISS. Keep intent and KISS as separate reviewers.
- Give every reviewer the same intent brief, repository rules, exact scope, changed files, and relevant diff.
Reviewers may run read-only checks and disposable probes but must not edit the candidate worktree.
- Verify every important finding yourself against current source and discard speculative or duplicate reports.
- For a final gate, run or inspect the relevant test matrix and confirm the reviewed head has not moved.
CreatioClient review invariants
- Existing synchronous contracts, exceptions, response bodies, and public members remain compatible unless a
deliberate breaking change is explicitly authorized.
- Do not add members to
ICreatioClientwithout treating third-party implementations as affected consumers. - Cookie, OAuth client-credentials, raw bearer, and NTLM modes retain their distinct login and renewal behavior.
- Credentials, cookies, CSRF values, and bearer tokens never cross an untrusted origin or appear in output.
- Redirects, retries, and authentication replay are bounded and recreate disposable request content safely.
- Cancellation reaches network and streaming operations; response and stream ownership remains explicit.
- The per-client pooled
HttpClientand authentication handler remain the single transport pipeline. - Tests use NUnit and FluentAssertions and cover
net8.0andnet10.0; live E2E is opt-in.
Report
Report findings first, ordered by Blocker, High, Medium, then Low. Each finding includes the lens, file and tight line range, trigger, evidence, impact, and smallest viable fix. Then report open questions that affect confidence, reviewed base/head, perspectives used, tests run or skipped, and residual risk. If there are no material findings, say so directly.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryChecking.ai/skills/agentic-code-review/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/advance-technologies-foundation/creatioclient/agentic-code-review)