cmd-pr-conflict-resolver

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes common developer tools such as git and ripgrep (rg) to inspect the current state of a repository, trace commit history, and identify code patterns. These commands are used for local analysis and are appropriate for the skill's functional context.
  • [DATA_EXFILTRATION]: While the skill reads repository data, including file contents and commit logs, it lacks any network-facing capabilities or instructions to transmit data to external servers.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from conflicting source files and git logs.
  • Ingestion points: File reads, git diff output, and git log output (SKILL.md sections 1, 2, and 3).
  • Boundary markers: Absent; there are no specific delimiters instructed for the agent to isolate untrusted data.
  • Capability inventory: Shell command execution (git, rg) and file modification/staging (git add).
  • Sanitization: None; the agent is expected to process the file content directly to resolve conflicts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 04:09 AM
Security Audit — agent-trust-hub — cmd-pr-conflict-resolver