anti-sycophant-ai-agent-skills

Reddit-only AI agent skill pack with anti-sycophancy prompts

A single Reddit post (#5 in a tiny 7-row subreddit pool) is the only detectable signal for this skill repository. No GitHub, HuggingFace, Hacker News, or developer platform traction exists.

Why now: LLM sycophancy is a discussed topic (Claude 3.5, GPT-4o criticism), so keyword matching may have triggered Reddit's algorithmic surfacing of any related content.

• Only 1 of 8 monitored sources (Reddit) shows any signal
• Reddit pool is minuscule: 7 rows total, rank 5 means near-bottom of near-nothing
• Zero GitHub stars/activity, zero HuggingFace presence, zero HN discussion
• Consensus confidence is 10/100 — lowest possible tier
• 175 of 200 pool items are single_source — this is statistical noise floor

Considerations: The 'anti-sycophant' framing is timely given LLM criticism, but timely framing without distribution is indistinguishable from content marketing. No code velocity, no forks, no issues, no citations. Could be a personal experiment that resonated with zero developers.

EMERGING SIGNAL · Ignore: Do not track; if signal appears on GitHub or HN with velocity, re-evaluate.

Sources: Reddit search: machinesoul11/anti-sycophant-ai-agent-skills