safe-implementer
Create or modify code, tests, configuration, and project files safely with real file changes.
- License 许可
- MIT
- 所属仓库
- domehahn/skcr
- 最近更新
- 2026-09-11
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- safe-implementer
- since
- 2026-06-10
- authors
- [ "platform-engineering" ]
- version
- 1.1.0
- replaces
- null
- changelog
- [ { "date": "2026-06-10", "change": "Added production-ready skill-specific operating model, checklist, decision rules, acceptance criteria, and output requirements", "version": "1.1.0" }, { "date": "2025-01-01", "change": "Initial release", "version": "1.0.0" } ]
- stability
- stable
- supersedes
- []
- description
- Create or modify code, tests, configuration, and project files safely with real file changes.
- last_modified
- 2026-06-10
- deprecated_since
- null
- min_platform_version
- { "kiro": "unknown", "codex": "unknown", "junie": "unknown", "cursor": "unknown", "ollama": "unknown", "opencode": "unknown", "roo-code": "unknown", "windsurf": "unknown", "openhands": "unknown", "gemini-cli": "unknown", "gitlab-duo": "unknown", "claude-code": "unknown", "github-copilot": "unknown" }
02 资源文件 6
- 参考文档CHANGELOG.md190 B
- 许可证LICENSE214 B
- 参考文档README.md339 B
- 参考文档skill.yaml393 B
- 参考文档tests/README.md92 B
- 其他VERSION6 B
03 SKILL.md 正文
Safe Implementer
Purpose
Create or modify code, tests, configuration, and project files safely with real file changes.
When to use
Use this skill when the task matches the description above or when the central agent instructions route work to $safe-implementer.
Operating model
- Clarify the goal and constraints.
- Inspect the minimum relevant repository context.
- Produce a concise execution plan for non-trivial work.
- Execute with tools when implementation is requested.
- Validate the result with repository-native checks.
- Summarize changed files, validation results, and residual risks.
DevSecOps guardrails
- Do not read secrets,
.envfiles, private keys, production credentials, masked CI/CD variables, database dumps, or sensitive logs unless explicitly required. - Do not push, deploy, publish, merge, or create releases unless explicitly asked.
- Prefer merge requests, reviewable diffs, and auditable validation evidence.
- Prefer least privilege, minimal changes, and explicit rollback notes.
- Do not fabricate test results, repository state, commands, or security findings.
Skill-Specific Operating Model
- Inspect existing patterns, ownership boundaries, tests, config, and generated files before changing code.
- Make the smallest coherent change that satisfies the request while preserving behavior outside scope.
- Protect user work: check the working tree, avoid reverting unrelated changes, and work with concurrent edits.
Skill-Specific Checklist
- Relevant code, tests, docs, config, and sync/generation paths were read before editing.
- Changes are scoped to the requested behavior and directly necessary support files.
- Formatting, tests, build, lint, or focused manual checks are run where practical.
Decision Rules
- Do not revert user changes unless explicitly requested.
- Do not stop at a plan when implementation is requested and context is sufficient.
- Ask before destructive filesystem changes, releases, deployments, or credential access.
Acceptance Criteria
- Requested behavior is implemented or the blocker is clearly documented.
- Relevant validation passes, or failures are reported with cause and next step.
- Generated or synchronized files match their canonical source when applicable.
Output Requirements
- State what changed and which files matter most.
- Report validation commands and results exactly as run.
- Mention residual risks, skipped checks, and follow-up work succinctly.
Changelog
1.1.0 - 2026-06-10
- Added production-ready skill-specific operating model, checklist, decision rules, acceptance criteria, and output requirements.
1.0.0 - 2025-01-01
- Initial release.
04 同仓库更多 Skills
vulnerability-management-coordinator
Assess CVE triage, prioritization, SLAs, exploitability, asset criticality, exceptions, risk acceptance, and remediation tracking.
verification-reviewer
Review diffs, validate acceptance criteria, inspect test results, and find missed requirements.
universal-skill-creator
Create, adapt, validate, and optimize reusable agent skills across agentic platforms.
threat-modeler
Identify assets, trust boundaries, abuse cases, attack paths, threats, and required security controls.
test-strategy-engineer
Design and generate unit, integration, regression, security, and end-to-end test strategies.
sre-reliability-reviewer
Assess SLOs, SLIs, error budgets, capacity, degradation, timeouts, retries, circuit breakers, load shedding, and operational risks.