测试设计、漏洞检测、质量保证（1.3万 个）
Check whether the MCP connection responds by sending a message and receiving it unchanged, or getting 'pong' when no message is supplied. Useful for CLI health checks, basic SignalR connection tests, and round-trip timing.
中文简介正在准备中。当前显示英语简介。
Run GDScript tests without opening the Godot editor and report failures through process exit codes that CI, an automated checking pipeline, can detect. Covers test runners, initial asset imports, asynchronous tests, and troubleshooting hangs or misleading success results.
中文简介正在准备中。当前显示英语简介。
Verify work before reporting it as complete by running fresh checks and reading their results. Match each claim to evidence, including test output, build results, bug reproduction, code changes, and requirement checks.
中文简介正在准备中。当前显示英语简介。
Develop behavior changes by first observing an expected test failure, then making the smallest passing implementation. Refactor with tests passing, and document preserved behavior and measured runtime whenever tests are consolidated.
中文简介正在准备中。当前显示英语简介。
Verify code changes before committing or shipping by running tests, builds, and checks against requirements. Use fresh command output to support completion reports, confirm bug fixes, and check work reported by other agents.
中文简介正在准备中。当前显示英语简介。
Audit changed code by tracing execution paths and matching them to existing tests. Show coverage gaps and test quality, then generate tests for high-priority gaps using the project's current framework, helpers, and naming conventions.
中文简介正在准备中。当前显示英语简介。
Reviews code for common security weaknesses, authentication problems, exposed credentials, and unsafe configuration. It adjusts review depth to the changes and offers an adversarial workflow, where defensive and attacking perspectives guide fixes and subsequent verification.
中文简介正在准备中。当前显示英语简介。
Audit an application or selected features with a structured checklist. Record evidence for broken behavior, incomplete implementations, and inconsistent patterns, then organize findings by severity and propose a prioritized repair plan.
中文简介正在准备中。当前显示英语简介。
Review code or documents with available AI providers before delivery. Combine findings into a scored validation report that checks prior requirements and decisions, separates critical issues from warnings, and recommends next steps.
中文简介正在准备中。当前显示英语简介。
Apply safeguards when analyzing external articles or posts. Validate URLs, limit fetched content, label source text as untrusted data, and clean analysis output so embedded instructions are treated as material rather than commands.
中文简介正在准备中。当前显示英语简介。
Verify an engine app without a real terminal and record the full interaction as video. Produce a summary page with a clickable timeline, pass or fail checks, and snapshots to review the tested behavior.
中文简介正在准备中。当前显示英语简介。
Assess external plugins, connectors, or APIs before adoption by comparing capabilities, authentication, costs, data freshness, safety, and fallback options. Produce readiness assessments that distinguish planned checks from observed trials, including memory providers and realtime voice services.
中文简介正在准备中。当前显示英语简介。
Prepare datasets, logs, and exports for external sharing by classifying fields, removing unnecessary data, and replacing identifiers with pseudonyms. Check the exact outgoing file for remaining personal data and secrets, then record sharing metadata.
中文简介正在准备中。当前显示英语简介。
Organize responses to dependency vulnerabilities, leaked credentials, and dependency license concerns in existing code. Assess whether vulnerable code can be reached, order containment steps, and define the evidence needed before closing the event.
中文简介正在准备中。当前显示英语简介。
Test changed behavior with adversarial scenarios: difficult inputs and failure conditions designed to expose problems. Record expected behavior, observed results, and pass or fail evidence, then hand identified fixes to the selected coding executor for correction and retesting.
中文简介正在准备中。当前显示英语简介。
Assess whether a change is ready by organizing build, test, type-checking, security, documentation, and CI results. Produce a pass, hold, or block verdict with specific evidence gaps, failed checks, and required follow-up.
中文简介正在准备中。当前显示英语简介。
Check rendered web pages, images, documents, and terminal interfaces using actual captures. Define screen sizes and states to cover, identify layout and text problems, and produce a scored verdict tied to the exact source revision.
中文简介正在准备中。当前显示英语简介。
Review an AI agent or automation before execution for malicious embedded instructions, excessive permissions, exposed secrets, dependency risks, and destructive actions. Receive evidence-based findings, action boundaries, and proposed remediation steps.
中文简介正在准备中。当前显示英语简介。
Prepare safety requirements before trials involving robots, 3D printers, IoT relays, or lab hardware. Organize hazards, sensor and camera checks, operator approval, simulated runs, and emergency stops while distinguishing planned safeguards from observed trial evidence.
中文简介正在准备中。当前显示英语简介。
Check whether observed evidence supports each claim that work is complete. Submit the claim, evidence excerpt, and goal to Jev, then report objections that need resolution; a lack of objections does not constitute approval.
中文简介正在准备中。当前显示英语简介。
Send code changes to Jev for advisory checks on authentication, tests, and migration risks. Receive findings for each file, a combined list of concerns, and the highest severity to inform a review without granting merge approval.
中文简介正在准备中。当前显示英语简介。
Ask Jev to review a command before execution for secrets, outbound sends, and the scope of its impact. Receive a result and rule explaining whether an additional hold or a recommendation against execution applies.
中文简介正在准备中。当前显示英语简介。
Build a threat model, a structured analysis of possible attacks, from an application’s architecture and data flows. Identify valuable assets, trust boundaries, attack scenarios, control owners, and tests that would demonstrate whether each defense works.
中文简介正在准备中。当前显示英语简介。
Audit code, frontend behavior, and agent traces for ignored errors, fallbacks that hide failures, and unsupported success reports. Findings identify locations, severity, user impact, and evidence, with repair handoffs based on reproduced failures.
中文简介正在准备中。当前显示英语简介。