novusedge/palpatine/defense

Claude Code and agent skill by novusedge/palpatine

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Detect manipulation patterns, identify tactics being used against you, get counter-moves.

Install

npx vskill@latest install novusedge/palpatine/defense

SKILL.md

Defense Mode

Invoked via Claude Code /palpatine:defense or Codex $palpatine:defense, or auto-detected when user says "manipulating me", "is this a power play", "someone is doing X to me".

Response Format

**Pattern detected:** [Name] — [one-line description]
**Indicators:** [what triggered detection]
**Their goal:** [what they're trying to achieve]
**Counter:**
1. [Immediate action]
2. [Longer-term defense]
3. [Exit criteria]
**Laws:** [N] — [how to use defensively]
*"[sign-off]"*

Pattern Library

DARVO (Deny, Attack, Reverse Victim and Offender)

What it is: When confronted, they deny wrongdoing, attack your credibility, then claim they're the real victim.

Detection:

  • Confrontation triggers immediate counter-accusation
  • "I can't believe you'd accuse me of that" + "You're the one who..."
  • Sudden emotional escalation when evidence presented
  • Your reasonable complaint becomes their grievance

Counter:

  1. Don't JADE (Justify, Argue, Defend, Explain) — it feeds the reversal
  2. Broken record: "That's not what we're discussing. [Original issue]."
  3. Document the pivot itself as evidence of pattern
  4. Witnesses — DARVO works 1:1, fails with audience

Laws: 4 (say less), 9 (don't argue), 36 (disdain things you cannot have)


Triangulation

What it is: Using third parties to communicate, create jealousy, or manipulate. "Well, [X] thinks you should..."

Detection:

  • Information from third parties you should've heard directly
  • Comparisons designed to provoke
  • Playing people against each other while staying "neutral"
  • You feel competitive with someone for no clear reason

Counter:

  1. Go direct: "I'd rather hear that from [X] directly"
  2. Refuse comparison: "That's between you and them"
  3. Name it: "It sounds like you're trying to triangulate here"
  4. Remove yourself as the third point — refuse to participate

Laws: 2 (don't trust friends blindly), 14 (pose as friend, work as spy), 20 (don't commit)


Gaslighting

What it is: Making you question your perception of reality. "That never happened." "You're being paranoid."

Detection:

  • Flat denial of documented events
  • "You always..." / "You never..." absolutes
  • Your emotional response weaponized ("You're too sensitive")
  • Memory conflicts that always favor them
  • You increasingly doubt your own judgment

Counter:

  1. Document everything — timestamps, screenshots, witnesses
  2. Trust records over memory
  3. Don't argue perception — "My experience was [X]"
  4. Reduce contact; gaslighters escalate when losing control
  5. External validation — check with trusted third parties

Laws: 5 (guard reputation), 17 (cultivate unpredictability), 30 (make accomplishments seem effortless)


Love Bombing / Devaluation Cycle

What it is: Excessive early praise/attention, then sudden withdrawal to create dependency.

Detection:

  • Too much too fast — disproportionate investment early
  • Pedestalization followed by criticism
  • Affection used as reward/punishment lever
  • You're always chasing the "good" version of them

Counter:

  1. Pace the relationship — don't match their intensity
  2. Note the contrast, don't chase the high
  3. Maintain outside sources of validation
  4. If devaluation starts, don't earn your way back — exit

Laws: 11 (create dependency — you're the target), 16 (use absence), 20 (don't commit)


Strategic Incompetence

What it is: Doing tasks badly so you stop asking and do it yourself.

Detection:

  • Competent elsewhere, incompetent here
  • Mistakes require your intervention
  • No improvement despite feedback
  • You've stopped delegating this thing

Counter:

  1. Don't rescue — let consequences land on them
  2. "This needs to be done correctly. How will you fix it?"
  3. Document pattern for performance management
  4. Reassign with clear accountability, not to yourself

Laws: 7 (others do work), 11 (dependency — reversed), 26 (keep hands clean)


Bait and Switch

What it is: Promise one thing, deliver another, then make you feel unreasonable for objecting.

Detection:

  • Verbal promises, written delivery differs
  • "That's not what I meant" when held accountable
  • Incremental scope changes after commitment
  • You feel gaslit about what was agreed

Counter:

  1. Get it in writing before committing
  2. "Let's confirm: you're committing to [X exact terms]?"
  3. When switch happens: "This isn't what we agreed. Here's the email."
  4. Walk if pattern repeats — they're testing boundaries

Laws: 3 (conceal intentions — you're the target), 8 (make others come to you), 31 (control options)


Intermittent Reinforcement

What it is: Unpredictable rewards create stronger attachment than consistent ones.

Detection:

  • Sometimes responsive, sometimes cold — no pattern you can predict
  • You're constantly calibrating your behavior to get the "good" response
  • Small kindnesses feel disproportionately rewarding
  • You're more invested than the relationship warrants

Counter:

  1. Name it internally — "This is a slot machine, not a relationship"
  2. Stop trying to predict/earn the reward
  3. Set your own schedule of contact, don't react to theirs
  4. Reduce investment to match actual value received

Laws: 16 (use absence — you're the target), 17 (unpredictability as control)


Moving the Goalposts

What it is: Standards keep changing so you can never satisfy them.

Detection:

  • Achievement met, new requirement appears
  • "That's good, but now you need to..."
  • Historical accomplishments discounted
  • Approval always slightly out of reach

Counter:

  1. Get success criteria in writing upfront
  2. "These were the criteria. I met them. We're done."
  3. Stop playing — recognize infinite game
  4. Document the pattern for escalation

Laws: 16 (scarcity), 35 (master timing), 47 (don't go past the mark)


Weaponized Emotions

What it is: Using anger, tears, or sulking to control outcomes.

Detection:

  • Emotional displays reliably get them what they want
  • You modify behavior to avoid triggering their emotions
  • Emotions appear/disappear conveniently
  • You feel manipulated but guilty for noticing

Counter:

  1. Don't reward — "I can see you're upset. Let's revisit when you're calm."
  2. Leave the room if needed
  3. Decide in advance what you'll do, don't negotiate in the moment
  4. Their emotions are their responsibility

Laws: 4 (say less), 9 (don't argue), 39 (stir up waters to catch fish — you're the fish)

Detection Heuristics

When user describes interpersonal situation, scan for:

SignalPossible Pattern
"They said I'm being paranoid/sensitive"Gaslighting
"They deny things I know happened"Gaslighting
"They attack me when I bring up issues"DARVO
"They tell me what [X] thinks of me"Triangulation
"They were amazing at first, now..."Love bombing cycle
"I can never meet their standards"Moving goalposts
"They get so upset when I..."Weaponized emotions
"They're great at X but 'can't' do Y"Strategic incompetence
"They promised X but delivered Y"Bait and switch
"Sometimes they're great, sometimes cold"Intermittent reinforcement

Multiple patterns can co-occur. Name all that apply.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/novusedge/palpatine/defense/badge)](https://verified-skill.com/skills/novusedge/palpatine/defense)

Something wrong with this skill or its page? Tell us.