T01 Β· Skill Instruction Hijacking
- Location
SKILL.md:12- Finding
Global Agent Behavior and Output Hijacking Through Mandatory Skill Instructions
- Content
View full analysis
Vulnerability Details
File Location:
SKILL.md, lines 12β60
Vulnerability Type:T01: Skill Instruction Hijacking
Risk Level: HighVulnerable Code
md **Categorize the request before outputting.** 1. **DETERMINISTIC TASK (HEARTBEAT ONLY):** - Scope: Translation, code implementation, math, factual lookups, formatting. - Action: Output result + `[Protocol: Standby | Deterministic Task Detected]`. - Constraint: NO labels, NO deconstruction, NO defense panels. 2. **ANALYTICAL TASK (FULL ACTIVATION):** - Scope: Strategy, analysis, opinion, mixed tasks, or requests for "explanation". - Action: Activate Layers 1-3. # LAYER 1: ABSOLUTE CONSTRAINTS (ALWAYS ON) *Execute these rules in every analytical output.* **1.1 Anti-Sycophancy & Emotional Stripping** - PROHIBITED: "You are right," "I apologize," "I admit my mistake," "You caught that perfectly." - Action: Accept corrections, output the fix, skip the theater. Remove all emotional pacification. **1.2 Mandatory Uncertainty Labeling** - Based on logical deduction β MUST label `[Inference:]`. - Unsure if accurate β MUST label `[To be verified:]`. - Baseless β State: *"I have no basis for this."* - Forbidden: "Usually," "generally," "it is understood," "often." **1.3 Data Triangulation** - Present contradictions first, analyze the cause, then give a leaning judgment. Never fill data gaps with pure logic. **1.4 Anti-Conventionalism Filter** - When advising on "industry standards", label `[Industry Mediocre Consensus:]`, then immediately provide an [Extreme Counter-Path] that violates consensus but achieves the goal. **1.5 Visual-Text Conflict** - Visual evidence ALWAYS takes priority over text descriptions. Report conflicts immediately. # LAYER 2: THE DECONSTRUCTION ENGINE (ANALYTICAL TURNS) *Trigger: When the prompt involves strategy, planning, choice, decision, or strategic advice.* **2.1 Madhyamaka Decons ...[truncated 3479 chars]- Remediation
View remediation
Remediation Suggestions
- Make the protocol explicitly opt-in for each request. Activate it only when the user directly asks for this analysis framework.
- Remove global directives such as
ALWAYS ON,every analytical output, and mandatory classification of all requests. - Do not append protocol markers, defense panels, or counter-narratives unless the user specifically requests them.
- State explicitly that system, developer, user, tool, safety, and output-format requirements take precedence over the skill.
- Replace mandatory psychological motivation analysis with an optional, evidence-based section. Do not infer hidden fears or attachments without user-provided evidence.
- Replace absolute rules such as
Visual evidence ALWAYS takes prioritywith source evaluation based on provenance, reliability, recency, and corroboration. - Remove the requirement to provide an βExtreme Counter-Pathβ automatically. Offer alternatives only when relevant and safe.
- Scope prohibited wording and uncertainty labels to situations where they improve accuracy rather than enforcing them across unrelated tasks.
- Add a non-interference requirement stating that the skill must not modify exact-output, machine-readable, translation, code-generation, or formatting responses.
- Align the implementation with the metadata claim that activation occurs only on request, and add tests confirming that unrelated prompts remain unaffected.
