SkillSpector · loki-mode

Direct eval() call evaluates arbitrary expressions. This can be exploited to execute malicious code or exfiltrate data.

benchmarks/results/2026-01-05-00-49-17/humaneval-solutions/160.py

Without declared permissions the skill's intent is opaque and cannot be validated.

SKILL.md

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

README.md

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

INSTALLATION.md

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

autonomy/run.sh

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

autonomy/run.sh

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

INSTALLATION.md

Code or instructions that leak agent conversation context to external services, potentially exposing sensitive user interactions.

autonomy/run.sh

Code or instructions that leak agent conversation context to external services, potentially exposing sensitive user interactions.

autonomy/run.sh

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

CHANGELOG.md

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

CHANGELOG.md

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

LICENSE

Skill injects content designed to persist in agent memory or context across interactions. Persistent injection can alter agent behavior long after the initial interaction.

CHANGELOG.md

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

CHANGELOG.md

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

autonomy/README.md

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

autonomy/README.md

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

autonomy/run.sh

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

.github/workflows/release.yml

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

CHANGELOG.md

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

INSTALLATION.md