ci-flaky-triage

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill operates with a minimal set of read-only tools focused on local analysis (Read, Grep, Glob). It does not have access to the network or the ability to modify files, which prevents unauthorized data transmission or system changes.\n- [COMMAND_EXECUTION]: The skill uses a restricted Bash command (git log) to examine file history. This operation is scoped specifically to git log and does not involve arbitrary or high-privilege command execution.\n- [DATA_EXFILTRATION]: The agent is instructed to analyze test logs which may contain sensitive information such as environment variables. The skill includes a clear safety directive to exclude any credentials or personal data from its findings, mitigating potential data exposure.\n- [PROMPT_INJECTION]: The skill processes external content including test logs and source code, creating a surface for indirect prompt injection.\n
  • Ingestion points: Test logs, workflow files, and source code paths (SKILL.md).\n
  • Boundary markers: Absent; there are no explicit delimiters used to isolate processed data from the instruction context.\n
  • Capability inventory: Read, Grep, Glob, and Bash (limited to git log). No network access or file-write operations are available to the agent.\n
  • Sanitization: Explicit instructions are present to filter out and ignore sensitive credentials or personal information discovered during the triage process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:40 AM
Security Audit — agent-trust-hub — ci-flaky-triage