The Agent Skills Directory

[SAFE]: No executable code, shell commands, or network operations are included in the skill. The logic is entirely text-based and instructional.
[NO_CODE]: The skill consists solely of Markdown documentation and does not define any Python or Node.js dependencies, nor does it attempt to download or execute remote code.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it reads untrusted user-provided documents, though the risk is mitigated by the lack of dangerous tool capabilities. 1. Ingestion points: Assignment materials including ToR, RFP, and advert briefs are ingested from the user's workspace (SKILL.md). 2. Boundary markers: The skill does not implement specific delimiters or instructions to the model to ignore potential malicious prompts embedded within the input documents. 3. Capability inventory: The skill is restricted to text analysis and drafting. It does not utilize any tools for network communication, file system writing, or arbitrary command execution. 4. Sanitization: No sanitization or filtering of external input is performed prior to the drafting process.

06-methodology