Security audit

Pilot Sla

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small, disclosed SLA-monitoring guide for pilotctl workflows, with one unsafe jq interpolation pattern users should harden before relying on enforcement results.

Install only in an environment where pilotctl task access is appropriate. Before using the SLA calculations for real penalties or compliance decisions, change the jq snippets to pass AGENT and TASK_ID via jq --arg, validate inputs, handle pilotctl/jq failures, and avoid dividing by zero when no matching tasks exist.

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: SKILL.md:62
Finding: Untrusted AGENT Value Injected into jq Expressions
Content: View full analysis
Remediation: View remediation
&2 exit 1 fi TOTAL=$(pilotctl --json task list | jq --arg agent "$AGENT" '[.[] | select(.target == $agent)] | length') SUCCESSFUL=$(pilotctl --json task list | jq --arg agent "$AGENT" '[.[] | select(.target == $agent and .status == "completed")] | length') ``` Apply the following additional hardening: 1. Validate `AGENT` against the identifier format accepted by `pilotctl`. 2. Check the exit status of both `pilotctl` and jq before using their results. 3. Confirm that `TOTAL` is numeric and greater than zero before division to prevent invalid success-rate calculations. 4. Apply the same `jq --arg` pattern to `TASK_ID` in the workflow example rather than embedding it into jq program text. 5. Ensure SLA enforcement decisions fail safely when task data cannot be retrieved or parsed. ]]>

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.