weekly-plan
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script located at
~/.claude/scripts/weekly-plan.shto gather activity data from Git and GitHub. The skill relies on this pre-existing script being present and executable on the host system. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted external sources, creating a surface where malicious instructions embedded in GitHub issues or repository files could influence the agent's behavior.
- Ingestion points: The skill reads daily log files, deadline-related files or comments in "grant repos," and external GitHub issues assigned to the user.
- Boundary markers: The instructions do not define specific boundary markers or delimiters to separate untrusted external data from the skill's core instructions.
- Capability inventory: The skill has the capability to execute shell scripts and write markdown files to the local filesystem at
~/code/claude-log/. - Sanitization: There is no mention of sanitization, filtering, or validation of the content retrieved from GitHub issues or repository comments before it is used for summarization and prioritization.
Audit Metadata