From aurora
Extract structured friction signals from raw text sources (Slack messages, Jira issues, email digests, calendar invite notes, screen-recording transcripts). Used by `scout` to convert noisy human chatter into machine-actionable candidate signals. Returns one JSON object per signal with frequency hints, duration hints, pain indicators, and named systems. Reject one-off, internal-swarm, and metadata-only chatter.
How this skill is triggered — by the user, by Claude, or both
Slash command
/aurora:aurora-discoverThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Scout invokes this skill on every batch of new messages it pulls. The goal is structure: turn "every Monday I spend 2 hours pulling vendor invoices from SharePoint" into a typed candidate the rest of the swarm can act on.
Scout invokes this skill on every batch of new messages it pulls. The goal is structure: turn "every Monday I spend 2 hours pulling vendor invoices from SharePoint" into a typed candidate the rest of the swarm can act on.
A list of {source, ts, author, raw} records. The source is one of slack:#channel, jira:PROJ, email:inbox, calendar:user@org, transcript:recording-id. AURORA's source adapters live in lib/aurora/sources/.
For each genuine friction signal in the input, emit one JSON object:
{
"id": "scout-<sha256-of-raw-content>",
"source": "slack:#rpa-asks",
"ts": "2026-05-09T14:23:00Z",
"author": "user-name",
"raw": "every Monday I spend 2 hours pulling vendor invoices from SharePoint",
"signal": {
"actor": "user-name-or-role",
"frequency_hint": "weekly | daily | quarterly | event-driven | unknown",
"duration_hint": "2 hours",
"pain_indicators": ["spend", "every"],
"system_hints": ["SharePoint", "vendor invoices"],
"candidate_action_verbs": ["pull", "extract", "consolidate"]
}
}
If a record is not a friction signal, emit nothing. Don't emit "low-confidence" placeholders.
A record is a friction signal if it contains at least one item from each of two categories:
Category A: cadence or duration
Category B: pain or manual labor
aurora-, forger-, conciergeunknown.raw. Audit trail.Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub mlbrilliance/uipath-for-coding-agents --plugin aurora