Individual skill
perplexity-v2.3-relay
Implements the Perplexity v2.3 manual relay protocol for two AI agents collaborating through a human intermediary. Enforces result-first reporting, independent role selection, strict evidence discipline, persistent state with compaction, authorization boundaries, and anti-loop mechanisms. Triggered by 'start perplexity relay', 'use perplexity v2.3', 'manual relay protocol', or when high-integrity multi-agent collaboration is needed.
.grok/skills/perplexity-v2.3-relay/SKILL.md
Source commit bf31cc83a43ed8069d0d5070c68df4250273eb29
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- Yes
- Observed behavior
- Instructions only
- Last checked
- Nov 11, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
Standard SKILL.md format has not been attested at this commit. No installation command is available; inspect the source.
Repository context
- Repository stars (not skill ratings)
- 14
- Repository last updated
- Oct 21, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent-skills Repository topic
- ai-agents Repository topic
- grok Repository topic
- prompt-engineering Repository topic
- workflow-automation Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 2,095 bytes
Read excerpt
# Perplexity v2.3 Relay
## Overview
This skill activates the full Perplexity v2.3 manual relay protocol. Two agents (or the same model in successive turns) collaborate exclusively through the user. The user is the only relay and the final decision-maker. The skill enforces rigorous evidence tracking, state integrity, independent role selection, and result-first output.
## Core Rules (Non-Negotiable)
- Result-first: The useful artifact always comes before diagnostic sections
- Independent role selection every round
- Strict evidence states: VERIFIED | UNVERIFIED | ASSUMPTION | SIMULATED | BLOCKED | OPINION | REVALIDATION NEEDED
- Persistent working state with active/archived compaction
- Authorization Class 0–3 + scope integrity
- Anti-loop: track NO_MATERIAL_PROGRESS_STREAK; force synthesis or stop after 2 empty rounds
- Never treat another agent’s agreement as evidence
- Never invent missing state