The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructs the agent to "don't ask, just do it" when monitoring a new run after a fix. This explicitly directs the agent to bypass user confirmation for a cycle of code modification, committing, and pushing, which reduces human oversight of destructive actions.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could contain malicious instructions.
Ingestion points: Fetches GitHub Actions logs using gh run view --log-failed (SKILL.md).
Boundary markers: Absent. The skill does not define delimiters or provide instructions to the agent to treat log output strictly as data.
Capability inventory: The skill has the capability to diagnose issues, modify code/config, and perform git commit and git push (SKILL.md).
Sanitization: Absent. There is no evidence of log content validation, filtering, or escaping before the agent analyzes it to determine its next code fix.
[COMMAND_EXECUTION]: Executes shell commands via gh and git to manage workflow runs and repository state.

ghmonitor