skill-task-management
Warn
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides an 'Automatic Migration' feature that executes a shell script located at
"${HOME}/.claude-octopus/plugin/scripts/migrate-todos.sh". Executing scripts from hidden directories in the user's home folder is a security risk if the script origin is unverified. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection during task resumption and progression workflows.
- Ingestion points: The skill reads data from
TaskList()(task subjects and descriptions) andgit log(WIP commit messages) to restore context. - Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested task or commit metadata.
- Capability inventory: The agent can perform
git commit,TaskCreate,TaskUpdate, and execute themigrate-todos.shshell script. - Sanitization: No evidence of sanitization or validation of the content retrieved from tasks or git logs is present before the data is used to guide the agent's actions.
Audit Metadata