The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external data, such as commit messages and PR descriptions, creating a potential surface for indirect prompt injection attacks where malicious instructions could influence the recap synthesis.
Ingestion points: Processes local activity logs from memory/logs/*.md and external data retrieved via the GitHub API.
Boundary markers: The instructions lack explicit delimiters or guidance for the agent to isolate or ignore potentially adversarial instructions embedded within the ingested data.
Capability inventory: The agent can write to the filesystem and execute shell commands via gh and git tools.
Sanitization: No specific sanitization or content validation logic is described for the retrieved external text before it is summarized.
[COMMAND_EXECUTION]: The skill uses command-line utilities to gather repository metrics and deliver notifications.
Evidence: Utilizes gh api, gh pr list, git, and a local ./notify script to perform its functional tasks. These operations are within the expected scope for a development-focused reporting skill.

operator-scorecard