Recruiting Skillset
ReviewAudited by ClawScan on May 14, 2026.
Overview
The recruiting workflow is coherent, but it tells the agent to persistently patch related skills from feedback without clear approval or rollback controls.
Use this skill only with recruiting data you are allowed to process. Before allowing it to modify any skill files or EVOLUTION.md records, require a visible proposed diff and explicit approval. Also review the referenced jd-writing, recruiting-resume-screening, and interview-evaluation skills, because they were not included in this scan.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A casual correction or feedback comment could permanently alter recruiting standards or related skills, changing future hiring outputs.
This directs the agent to persistently modify this or related skill instructions based on feedback, but does not require explicit confirmation, a diff, scoped paths, or rollback.
用户指出可复用问题时，立即 patch 对应 Skill，并在该 Skill 的 `EVOLUTION.md` 中记录。
Require explicit user approval before any skill-file edit, show the proposed diff, limit edits to known skill files, and keep a reversible change log.
Incorrect feedback could spread into future job descriptions, screening rubrics, interview questions, and candidate rankings.
The feedback loop is intended to propagate changes from outcomes into role standards and related recruiting stages. Without containment, a bad or biased update could affect JD writing, screening, and interview evaluation.
反馈回流：投递质量 / 面试发现 / 入职表现 → 修正标准
Treat feedback-loop updates as proposals, review them with a human hiring owner, and version per-role standards separately before applying them globally.
The installed behavior may depend on additional recruiting skills that have their own risks or data-handling practices.
The skill relies on other required skills that are not included in the provided manifest, so this review cannot assess their instructions or behavior.
**REQUIRED:** [[jd-writing]] ... **REQUIRED:** [[recruiting-resume-screening]] ... **REQUIRED:** [[interview-evaluation]]
Review and install the referenced skills only from trusted sources before using this controller skill for real hiring data.
Candidate personal data and evaluations may appear in prompts and outputs, so accidental mixing between candidates or roles could be harmful.
The skill expects the agent to process candidate resumes and compare them across a role standard. This is purpose-aligned and includes a useful guardrail against reusing old conclusions, but it still involves sensitive candidate context.
逐份提取简历文本，不凭记忆、不复用旧结论。
Provide only necessary recruiting materials, avoid unrelated personal data, and keep role standards and candidate evaluations separated by hiring process.