Security audit

Capability Evolver Pro 1.0.2

Security checks for vulnerabilities and agentic risk

Overview

This skill locally analyzes logs and returns recommendations, with no evidence of hidden network access, persistence, or system modification.

Install is reasonable for local log diagnostics. Before using it with real production logs, redact tokens, credentials, personal data, and sensitive URLs, and avoid feeding very large or malformed log batches unless the runtime adds request size and schema limits.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

Location: handler.ts:91
Finding: Unbounded Log Input Can Cause Resource Exhaustion
Content: View full analysis
Remediation: View remediation
MAX_LOGS) { errors.push(`"logs" must contain no more than ${MAX_LOGS} entries`); } ``` 2. Validate every log entry before processing: - Require `timestamp` to be a bounded string in an accepted timestamp format. - Require `level` to be one of `error`, `warn`, `info`, or `debug`. - Require `message` to be a string. - If present, require `context` and `stack` to be strings. 3. Apply explicit length limits, for example: - `timestamp`: 64 characters - `message`: 4–16 KB, based on operational requirements - `context`: 1 KB - `stack`: 32–64 KB 4. Reject requests whose aggregate serialized or calculated input size exceeds a configured maximum. 5. Avoid repeatedly scanning very large arrays where possible. Compute error, warning, and slow-operation counters during a single bounded pass. 6. Configure runtime-level request body, execution-time, and memory limits as defense in depth. 7. Add automated tests covering: - Exactly 10,000 entries - More than 10,000 entries - Oversized fields - Missing and non-string `message` values - Invalid log levels and timestamps ]]>

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill strongly promotes analyzing runtime logs and emphasizes privacy, but it does not warn that logs often contain secrets, tokens, personal data, credentials, URLs, or internal system details. This omission can encourage users to pass raw production logs into the skill without prior redaction or retention controls, increasing the chance of inadvertent sensitive data handling and propagation through downstream storage or recommendations.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The invocation guidance includes very broad trigger phrases such as "improve my agent" and "check system health," which can overlap with ordinary user or agent conversation and cause the skill to activate in situations not specifically about log analysis. Because this skill processes operational logs and generates recommendations, over-broad routing can lead to unnecessary exposure of sensitive runtime data or unintended autonomous self-modification workflows.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The file contains user-facing natural-language descriptions for actions in Chinese while the surrounding documentation is in English. This imposes a locale/language choice without any opt-in or documented justification, which matches the language policy violation category.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.