The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill connects to X (Twitter), a well-known service, to retrieve bookmarks and authored posts for analysis.
Source: https://x.com (including bookmark feeds and search timelines) as specified in SKILL.md.
Action: Uses the browser to collect content for voice calibration and drafting.
[COMMAND_EXECUTION]: The skill utilizes standard version control commands to manage the generated content as directed in SKILL.md.
Tools: git status, git diff, and git commit.
Context: These are used to stage and commit new quote-post drafts to the local workspace.
[PROMPT_INJECTION]: The skill processes untrusted external data from X bookmarks, creating a surface for indirect prompt injection.
Ingestion points: External tweet text and metadata are collected from X.com via browser instructions in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or warnings to separate external data from system instructions.
Capability inventory: The skill possesses the capability to write markdown files to the local data/ directory and execute git commit to persist changes as outlined in SKILL.md.
Sanitization: No specific content filtering or sanitization of the retrieved external text is performed before it is processed by the model.

x-bookmark-quote-posts