HR sensitivity — preserve evidence, not verdicts

this page documents synthetic reference data and local-only tooling — not HR advice, employment law guidance, or a substitute for your organization's investigation policy. confirm you are authorized under company policy before collecting chat exports; loop in HR and employment counsel on scope — especially when the accused holds management authority over the complainant. preserve originals read-only, hash every file sha-256 before redacting, and do not treat stylometry as a termination decision. counsel and HR determine findings and remedies.

• trigger a legal hold on slack, teams, and mailbox exports before channels purge or retention expires.
• collect from employer-owned systems and authorized exports — not personal devices without policy and counsel approval.
• if the complainant requests confidentiality, route sharing through HR — not manager chains that include the accused.

what this proves

• all eight workplace-harassment primary engines produce deterministic, fixture-locked output — verified by npm run check:flagship (280/280 fleet · 8 for this scenario).
• every output is generated 100% locally in your browser — export slack, teams, and email artifacts offline, never upload HR data.
• deleted-message subtypes, teams edit/delete timestamps, desktop sqlite residue, threaded hostile email, and anonymous-post authorship similarity surface without sending evidence to a server.

primary engines locked to this fixture

• slack-export-analyzer
• slack-export-forensic-analyzer
• teams-export-forensic-analyzer
• slack-forensics
• teams-forensics
• email-thread-reconstructor
• email-threading-reconstructor
• writing-sample-authorship-comparator

build the case binder

runs all eight primary engines on the synthetic evidence zip and opens a self-contained html binder. uses the default binder renderer for workplace harassment — no upload.

download the synthetic evidence

MIT-licensed, fully synthetic. includes slack workspace export zip, teams messages/channels/members json, hostile and follow-up .eml thread PKR-HR-THREAD-001, accused and anonymous writing samples, slack storage sqlite, and teams cache sqlite.

built deterministically from scripts/fixtures/build-parker-workplace-harassment.mjs. seed: parker-workplace-harassment:v1.

methodology

workplace harassment is a pattern across platforms, not one rude slack message. start with legal-hold slack and teams exports, then walk deletion metadata, endpoint cache residue, email thread continuity, and authorship comparison — D. Mitchell's voice in #hr-policy, edited teams posts, and PKR-HR-THREAD-001 email tie the cluster before stylometry supports anonymous authorship. read the full workplace harassment / hostile workplace guide →

after the playbook

run each primary locally — or export findings from the binder — then drop every csv/json into fatcousin-multi-tool-super-timeline-correlator. one timestamp-sorted timeline across slack deletions, teams edits, PKR-HR-THREAD-001 email, and authorship similarity — still zero upload.