Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward paid Botlington audit integration that sends user-provided agent details to Botlington, with no hidden code, persistence, or local privilege behavior found.
Install only if you are comfortable sending agent configuration, prompts, schedules, tool lists, and audit answers to Botlington's external service. Treat the API key as a secret, use an environment variable or secrets manager, and redact internal identifiers, credentials, personal data, and regulated information before submitting an audit.
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
Stream results as they arrive:
curl -N "https://botlington.com/a2a?taskId=TASK_ID"
Events:
The skill explicitly routes agent configuration and conversational inputs to Botlington's third-party A2A endpoint, and the description does not clearly warn users that potentially sensitive operational details will leave their environment. Because the example payloads include cron schedules, prompts, tools, and context strategy, users could disclose internal architecture or sensitive business data without informed consent.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
Stream results as they arrive:
curl -N "https://botlington.com/a2a?taskId=TASK_ID"
Events:
This POST transmits user-provided audit answers and session identifiers to a third-party endpoint, which can include sensitive details about internal cron jobs, workflows, and tooling. In the context of an agent-auditing skill, that external transmission is central functionality, but it still creates a real confidentiality risk if users are not warned or if secrets are included in responses.
echo "Gary: $QUESTION"
# 2. Answer Gary's question
curl -s -X POST $BASE \
-H "Content-Type: application/json" \
-H "x-api-key: $API_KEY" \
-d "{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tasks/send\",\"params\":{\"id\":\"$TASK_ID\",\"message\":{\"role\":\"user\",\"parts\":[{\"kind\":\"text\",\"text\":\"I run 8 cron jobs, every 15-60 minutes.\"}]}}}"
The API key is described as something to obtain and set in headers or the environment, but there is no explicit warning that it is a secret credential that must not be hardcoded, logged, or shared. This omission increases the chance of accidental exposure through shell history, code samples, screenshots, or repository commits.