PR Comment Politeness Evaluator
Assess whether a PR review comment sounds polite to foster a friendly collaboration environment.
The skill file itself only states to judge politeness, with no explicit permissions, confirmation, isolation, or rollback mechanisms; however, the repo has a SECURITY.md disclosing security policy, and dependencies explicitly exclude known vulnerable versions (e.g., GHSA-4xgf-cpjx-pc3j), indicating some dependency security awareness. Also, publisher identity is unverified, lacking clear maintenance responsibility. Deductions: skill mentions no data-flow transparency, user confirmation, or least privilege; repo-level security measures do not substitute for skill-level trust design.
Skill description is simplistic, with no instructions or scripts to execute, so key paths cannot be verified. Only a single-purpose statement exists without any input/output examples or error handling. Deductions: no reproducible tests or error feedback; static review cannot assess runtime reliability.
Use case is clear (politeness judgment), but lacks boundary definitions (e.g., language, context types) and precise trigger conditions. Repo is English-only; no Chinese support info and no declared overseas service dependencies, potentially affecting mainland users. Deductions: boundaries unclear, environment fit evidence insufficient.
SKILL.md is minimal, lacking installation, dependencies, examples, FAQ, or known-limitation disclosures; repo has Apache-2.0 license and version (pyproject.toml), but the skill file itself has no version or update path. Deductions: incomplete information architecture, maintenance responsibility unclear.
Skill claims to judge comment politeness but provides no output format or verification evidence; marginal value unknown. Deductions: value claim unverifiable, output usability evidence absent.
No tests or third-party execution evidence for the skill; only brief description. Repo has CI workflows but they target MCP server, not this skill. Deductions: key claims lack verifiable evidence.
- The skill file is extremely minimal, lacking input/output specifications, error handling, and boundary conditions, so real-world use may yield unreliable results.
- The repository is English-centric, with no Chinese documentation or offline support; mainland China users may experience service unreachability.
- Publisher is unverified; confirm maintenance responsibility and update path before adoption.
What does this skill do, and when should you use it?
This is a minimal Agent skill that evaluates the politeness of GitHub pull request review comments. It focuses solely on whether the comment sounds polite, ignoring code risk or technical correctness. The skill consists of a single SKILL.md file with a concise description and no accompanying scripts or configuration. It's ideal for teams wanting to automate quality checks on review communication and ensure positive interactions.
The skill reads a PR review comment and determines its politeness based on linguistic cues. It performs no code analysis, runs no external tools, and makes no network calls; it relies purely on the model's language understanding to produce a subjective judgment.
- In a code review workflow, automatically flag impolite comments to alert reviewers to adjust their tone.
- When maintaining an open-source project, monitor the tone of community contributions and intervene in conflicts early.
- As part of a feedback loop, pre-evaluate and polish comments generated by a model if they seem harsh.
- In large teams, standardize a respectful review culture and encourage constructive feedback.
What are this skill's strengths and limitations?
- Simple and direct, no dependencies, easy integration.
- Focused on a single task with clear outputs.
- Helps maintain a positive collaboration atmosphere.
- Judgment is subjective and may vary by culture or context.
- No scripts or rules, relies entirely on model judgment.
- No tests or benchmarks provided, accuracy unverified.
How do you install this skill?
Fetch the SKILL.md file from the repository and place it in your agent's skills directory (e.g., Claude Code's skills folder). The skill has no dependencies or extra installation steps.
How do you use this skill?
Trigger it via natural language in an agent-compatible environment. For example: "Evaluate this PR comment for politeness: 'Your code sucks, rewrite it.'" The skill returns a politeness verdict.