skills/discuss/SKILL.md
Use when a topic benefits from multiple perspectives debating before a decision.
npx skillsauth add kangig94/coral discussInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Start a backend-managed structured discussion with AI agents.
Pass --user to participate as a human observer.
If no topic is provided: gather it interactively, plus:
Analyze the topic and prepare persona inputs.
demographics: { origin_weights, outlier_ratio }.Generate persona seeds with:
coral-cli discuss seed --input-json -
using the prepared { controversy_axes, n, demographics?, seed } payload on stdin.
Turn the seed output into personas.
suggested_origin / is_outlier fields.Build the agent list and start the session.
{ name, persona, provider?, model? }.--user, add:
{ name: 'user', persona: '# User — Human Participant\nHuman observer with real-time participation via /bid skill.', participation: 'observer' }coral-cli discuss start --input-json - for the full { topic, agents, config? } payload.
Include config: { min_bid_delay_ms: 10000 } for --user; omit config otherwise.Session started: <session_id>.If --user, write CORAL_PROJECT/discuss/active-user-session.json
with { session_id }.
This keeps /bid pointed at the active observer session.
Monitor progress by polling:
coral-cli discuss watch --session "<session_id>"
cursor.Cursor: <n> value printed in the watch output:
coral-cli discuss watch --session "<session_id>" --cursor <previous_cursor>If --user, return immediately:
"Discussion started! Use /bid <score>, <thought> to submit a bid, or /bid <your speech> when you win the floor."
When the session ends, report the end reason from discuss_watch.
Session end is surfaced in the watch status line.
--user, delete CORAL_PROJECT/discuss/active-user-session.json.From the user’s request, identify:
--user mode--hintsdata-ai
Use when implementing a plan or executing a prompt that requires verified completion.
tools
Use when a problem needs clarification and agreement before planning begins. Supports --deep and --delegate.
tools
Use when a task needs structured planning before implementation. Supports --delegate and round=N.
data-ai
Use after implementation to run tier-classified review agents and produce a consolidated verdict.