The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local markdown files that could contain malicious instructions designed to influence the agent's output.
Ingestion points: The skill reads 01-job-analysis.md and 02_achievement_library.md from the user's home directory during the Phase 0 startup sequence.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when loading the file contents into the context.
Capability inventory: The skill possesses the ability to read and write to the filesystem and execute shell commands via the agent.
Sanitization: The risk is mitigated by explicit user confirmation steps in Phase 3 (Match & Select) and Phase 4 (Authenticity Validation), ensuring the user reviews and approves all tailored content before saving.
[COMMAND_EXECUTION]: The skill uses shell commands to facilitate file discovery and selection.
Evidence: Phase 0 specifies the use of ls ~/career-applications/*/01-job-analysis.md | sort -r | head -1 to identify the most recent job analysis file. This is a standard utility command for the skill's file-based workflow.

resume-alignment