The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from git logs, ticket descriptions, and specification files, which are external and potentially untrusted sources.
Ingestion points: Git history logs, issue tracker responses, and file system content (specs) referenced in SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the retrieved data.
Capability inventory: The skill relies on standard read-only commands (git log) and file searches to aggregate information. No destructive file operations or network exfiltration tools are used.
Sanitization: No explicit data sanitization or validation steps are defined for the ingested content.
[COMMAND_EXECUTION]: The skill instructs the agent to execute git commands to retrieve commit history for report generation.
Evidence: The workflow step 2 explicitly uses git log --since="{period}" --oneline --no-merges to gather data.

stakeholder-update