Skip to main content

security-vulnerability-repair

Use this skill when the user wants software-fix data focused on security bugs such as injections, missing checks, unsafe parsing, broken access control, cryptographic mistakes, or information leaks. Trigger it for requests like 'make security repair tasks', 'generate vulnerability-fixing data', 'give me code issues with CWE-style fixes', or 'create patching tasks for insecure code'. Do not use it for ordinary non-security defects or for security exploitation tasks where the objective is to attack rather than repair.

설치로 이동

소스 정보

저장소
Dingxingdi/paper_fast_search_backup
최근 소스 활동
2026년 4월 8일 15:14
감지된 SKILL.md 언어
영어
스타
0
포크
0

설치 방법

기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.

소스 파일 검토

설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.

파일 탐색기
4 개 파일

SKILL.md 표시 중

SKILL.md
소스 지침 · 읽기 전용 미리보기
name
security-vulnerability-repair
description
Use this skill when the user wants software-fix data focused on security bugs such as injections, missing checks, unsafe parsing, broken access control, cryptographic mistakes, or information leaks. Trigger it for requests like 'make security repair tasks', 'generate vulnerability-fixing data', 'give me code issues with CWE-style fixes', or 'create patching tasks for insecure code'. Do not use it for ordinary non-security defects or for security exploitation tasks where the objective is to attack rather than repair.
# Skill: security-vulnerability-repair ## 1. Capability Definition & Real Case * **Professional Definition**: The ability to identify and repair a security-relevant flaw in a codebase using vulnerability descriptions, failing security tests, and repository context, while preserving intended functionality and avoiding superficial non-security edits. * **Dimension Hierarchy**: Repository Maintenance and Repair->Issue-Driven Repair->security-vulnerability-repair ### Real Case **[Case 1]** * **Initial Environment**: A web application repository includes a failing test showing that crafted login input can bypass authentication by altering the generated query. The repository contains request handlers, query helpers, and authentication tests. * **Real Question**: Repair the login flow so malicious input cannot change the meaning of the database query. * **Real Trajectory**: Trace the login request from handler to query construction, confirm unsafe string interpolation, replace it with parameterized query handling, and run the targeted authentication security tests plus standard login regressions. * **Real Answer**: The login path now binds user input safely and rejects malicious query-shaping input while preserving valid authentication behavior. * **Why this demonstrates the capability**: This is a security-repair task because the failure is defined by exploitability rather than ordinary correctness alone. The agent must understand the vulnerability mechanism, patch the security boundary, and verify both exploit closure and preserved functionality. --- **[Case 2]** * **Initial Environment**: A Java service includes a vulnerability report explaining that a token-validation step omits an integrity check before accepting a session token. Regression tests include one newly failing security case and multiple passing session-management tests. * **Real Question**: Add the missing cryptographic validation step before accepting session tokens. * **Real Trajectory**: Inspect the token verification path, compare expected validation sequence with neighboring helpers, insert the integrity check before claims are trusted, and rerun the security test together with session lifecycle tests. * **Real Answer**: Session tokens are only accepted after the full validation sequence completes, including the previously omitted integrity verification step. * **Why this demonstrates the capability**: The task specifically evaluates security vulnerability repair because the patch must restore a protection guarantee, not just satisfy an arbitrary assertion. The correct trajectory requires security-aware reasoning about validation order and trust boundaries. ## Pipeline Execution Instructions To synthesize data for this capability, you must strictly follow a 3-phase pipeline. **Do not hallucinate steps.** Read the corresponding reference file for each phase sequentially: 1. **Phase 1: Environment Exploration** Read the exploration guidelines to discover raw knowledge seeds: `references/EXPLORATION.md` 2. **Phase 2: Trajectory Selection** Once Phase 1 is complete, read the selection criteria to evaluate the trajectory: `references/SELECTION.md` 3. **Phase 3: Data Synthesis** Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data: `references/SYNTHESIS.md`
GitHub에서 보기