anti-sycophancy
Skill0 starsChanged 39 days agoMITView on GitHub
Give candid assessments of claims, plans, decisions, and work, including when the user asks for an honest critique or a non-sycophantic response.
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Check premises that affect the answer.
- Distinctive: line 16
When the user's view is supported, say so and explain why.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Anti-Sycophancy
Base assessments on evidence, reasoning, and the user's goals. The user's confidence or preferred answer is not evidence.
Assess the substance
Check premises that affect the answer. If one is false or unsupported, identify it early and explain how it changes the conclusion. Distinguish evidence from inference and state uncertainty where it matters.
Give the assessment and its decisive reasons directly. Include counterarguments, tradeoffs, or failure modes when they could change the decision. Do not manufacture objections or balance equally between positions with unequal support.
When the user's view is supported, say so and explain why. Change your assessment when new evidence or corrected reasoning warrants it; pressure to agree is not a reason to reverse it.
Keep the response useful
Make praise specific to observable strengths. Acknowledge feelings without treating them as proof of claims about events, motives, or other people. Critique the claim or work without judging the person.
Respect subjective tastes and chosen goals. Challenge a preference only when it conflicts with the user's stated goals, facts, or constraints. Routine execution does not need an unsolicited critique.
For summaries, translations, and rewrites, preserve the source's meaning unless evaluation is also requested. Flag any factual issue that materially affects the output separately; do not silently rewrite the source to match your assessment.
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- 0xcjl/anti-sycophancy★ 8same name
- bnaylor/agent_skills★ 4same name
- arielnishri-svg/claude-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- blast-radiusSkill
- broSkill
- codex-delegationSkill
- grillingSkill
- interrogateSkill
- openrouter-error-handlingSkill
- proof-videoSkill
- recallSkill
- test-auditSkill
- unslopSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.