The Agent Skills Directory

[COMMAND_EXECUTION]: The skill contains instructions for the agent to use standard command-line tools such as git, awk, sed, sort, and tail to perform version checks and maintenance tasks.
[EXTERNAL_DOWNLOADS]: The skill is configured to fetch update tags and synchronized content from the author's official GitHub repository (Agents365-ai/365-skills). This process is transparent and requires human-in-the-loop confirmation.
[DYNAMIC_EXECUTION]: The skill features a self-update capability, allowing it to pull new instructions and configurations from its source repository via git pull to stay current with funding guidelines.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external content (user grant proposals).
Ingestion points: User-provided grant ideas, project titles, and draft logic are processed through the instructions in SKILL.md.
Boundary markers: The instructions use structured sections to separate reasoning layers, though they do not include explicit XML-style delimiters for user input.
Capability inventory: The agent is permitted to execute git for repository management and shell utilities for version comparison as described in the Update check section of SKILL.md.
Sanitization: The skill relies on the requirement for explicit user consent before executing commands that modify the local skill directory.

grant-thinking-cn-biology