The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation instructs users to install external dependencies using the command npx skills add skinnyandbald/ceos. These resources are hosted on GitHub and provided by the same author as the skill itself.
[COMMAND_EXECUTION]: The skill utilizes Bash for filesystem operations and data processing. It also references a specific hardcoded absolute path (/Users/ben/code/dear-ben/content/episodes/) for scanning local content, which may not be applicable or accessible in all environments.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes data from external, potentially untrusted sources.
Ingestion points: Reads data from Gmail message bodies, Google Calendar event details, Attio CRM records, and local L10 meeting notes (02_Areas/eos/data/meetings/l10/).
Boundary markers: The logic does not specify the use of delimiters or instructions to the agent to ignore embedded prompts within the ingested data.
Capability inventory: The skill has the ability to write to the local filesystem (Write tool) and execute shell commands (Bash tool).
Sanitization: While no programmatic sanitization is explicitly described, the skill implements a mandatory human-in-the-loop validation step. It uses AskUserQuestion to present every metric and its source to the user for sequential confirmation before any data is committed to disk.

eos