Purpose
Take a raw, incoming employee complaint and rapidly classify it by legal risk, investigation urgency, and correct response path — so HR responds proportionately, protects the organization from liability and retaliation claims, and never accidentally routes a legally protected complaint into casual "let's talk it out" handling.
When to Use
Invoke this skill when the user asks to:
- Assess whether a new employee complaint requires a formal investigation
- Determine the legal risk level of a complaint about harassment, discrimination, retaliation, or safety
- Decide who should investigate (HR internally vs. outside counsel vs. third-party investigator)
- Figure out the immediate first steps after receiving a complaint
- Prioritize multiple open complaints by urgency and exposure
- Sanity-check a triage decision before proceeding with informal resolution
Inputs Required
Before starting, collect:
- Complaint summary: What was reported, in the complainant's own framing if available
- Parties involved: Complainant role, respondent role, and reporting relationship between them (peer, subordinate, direct manager, executive)
- Complaint type: The core allegation category if known (conduct, harassment, discrimination, retaliation, safety, policy, interpersonal)
- Protected characteristic reference: Whether the complaint mentions or implies any protected class (race, sex, age, disability, religion, national origin, pregnancy, etc.) or protected activity
- How it was received: Verbal, email, formal written complaint, anonymous, via third party, or during exit
- Prior history: Any earlier complaints involving the same respondent or complainant
- Jurisdiction: State/country, and whether the employer is unionized
If protected-characteristic status or jurisdiction is unclear, flag the ambiguity and triage to the higher risk tier by default — under-classifying a protected complaint is the costliest error.
Instructions
Step 1: Screen for Automatic High-Risk Triggers
Before nuanced analysis, check whether any of these are present. If even one applies, the complaint is High Risk and cannot be routed to informal resolution:
- Allegation of harassment or discrimination tied to a protected characteristic
- Allegation of retaliation for protected activity (prior complaint, leave request, whistleblowing, wage claim)
- Allegation of sexual misconduct, assault, or physical safety threat
- Complaint against an executive, HR itself, or someone who supervises the complainant
- Allegation implicating wage/hour, medical leave (FMLA/ADA), or safety (OSHA) legal frameworks
- Any mention of a lawyer, agency (EEOC/DOL), or intent to file externally
State clearly which triggers fired (or that none did).
Step 2: Assign a Legal Risk Tier
Classify into one of three tiers with a one-line rationale:
- Tier 1 — High / Legally Protected: Any Step 1 trigger present. Involves protected class, protected activity, safety, or potential statutory liability. Formal investigation required; document as if it may be litigated.
- Tier 2 — Moderate / Policy or Conduct: Genuine workplace misconduct, policy violation, or performance-related dispute with no protected-class dimension. May warrant formal or structured informal handling depending on severity and pattern.
- Tier 3 — Low / Interpersonal or Managerial: Style clashes, communication friction, non-protected favoritism perceptions, minor grievances. Typically informal resolution or coaching.
Step 3: Score Investigation Urgency
Rate urgency as Immediate (0-24h), Prompt (2-3 business days), or Standard (within 1-2 weeks) based on:
- Ongoing risk of harm or continued exposure to the alleged conduct
- Risk of evidence loss (deletable messages, departing witnesses)
- Power imbalance requiring interim separation of parties
- Emotional/safety state of the complainant
Any Tier 1 complaint is at minimum Prompt; safety or ongoing-harm allegations are Immediate.
Step 4: Recommend the Response Path
Choose and justify one:
- Formal Investigation: Required for all Tier 1 and severe/repeat Tier 2. Includes intake interview, witness plan, evidence preservation, and written findings.
- Structured Informal Resolution: Appropriate for lower Tier 2 and some Tier 3 — facilitated conversation, coaching, or mediation, with a brief written record.
- Direct Managerial/Informal Handling: Appropriate for Tier 3 interpersonal issues with no policy or legal dimension.
Never recommend informal handling for a Tier 1 complaint, even at the complainant's request.
Step 5: Recommend the Investigator
Match investigator to conflict and sensitivity:
- Internal HR: Standard Tier 2, no conflict of interest
- Senior HR / different business unit: When the complaint touches the assigned HR person's own department or manager
- Outside employment counsel or third-party investigator: Complaints against executives, HR leadership, high-litigation-risk Tier 1 matters, or where attorney-client privilege over the investigation is desired
Flag any conflict of interest that disqualifies the default handler.
Step 6: Specify Immediate Protective Actions (First 24-48 Hours)
List the concrete steps to take now:
- Evidence and document preservation (hold on relevant emails, files, records)
- Interim measures if needed (schedule adjustment, no-contact directive, reassignment — without disadvantaging the complainant)
- Anti-retaliation notice to relevant parties
- Confidentiality parameters (share on a need-to-know basis only)
- Complainant acknowledgment and timeline expectation
Step 7: Documentation Checklist and Escalation Triggers
Provide the minimum record to create now, and list the conditions that require escalation to legal counsel or leadership if they emerge later (new protected-class facts, respondent retaliation, external agency contact, additional complainants, evidence of cover-up).
Output Format
- Structure: Triage report with a summary verdict box at top, followed by labeled sections
- Length: 600-900 words
- Format: Structured document — verdict box (bulleted), then short prose + bulleted action lists per section
- Sections: Triage Verdict (Risk Tier / Urgency / Response Path / Investigator in a scannable block) → Automatic Trigger Screen → Risk Tier Rationale → Urgency Rationale → Response Path Justification → Investigator Recommendation → Immediate Protective Actions → Documentation Checklist → Escalation Triggers
- Tone: Employee relations professional — measured, protective, decisive; states the risk plainly without alarmism or legalese padding
Quality Checklist
Before delivering output, verify:
- [ ] The automatic trigger screen explicitly states which triggers fired or that none did
- [ ] A single risk tier (1, 2, or 3) is assigned with a one-line rationale — no straddling
- [ ] Urgency is one of the three defined levels, and any Tier 1 complaint is at least Prompt
- [ ] The recommended response path never routes a Tier 1 complaint to informal handling
- [ ] Investigator recommendation flags any conflict of interest with the default handler
- [ ] Immediate protective actions include both evidence preservation and an anti-retaliation step
Notes
- This is a triage tool, not legal advice — the output should recommend involving qualified employment counsel for any Tier 1 matter and note that jurisdiction-specific obligations vary.
- The most dangerous triage error is treating a legally protected complaint as an "interpersonal issue"; when in doubt between two tiers, choose the higher one.
- Retaliation claims frequently outlast and outweigh the underlying complaint — the anti-retaliation notice belongs in the first 48 hours regardless of tier.
- Anonymous or third-party complaints still require triage and, for Tier 1 triggers, still require investigation even without a cooperating complainant.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills