The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements an indirect prompt injection surface through its 'Vector Consistency Stabilizer' component which parses external data for integrity checks.
Ingestion points: The extract_fact_atoms function in scripts/stabilizers/vector_consistency_check.py reads and extracts paragraphs and list items directly from markdown files.
Boundary markers: The format_report function includes the extracted fact content in a summary report without using delimiters or instructions to the agent to ignore potential embedded commands.
Capability inventory: The skill has the capability to read local filesystem files and perform semantic search queries against a vector database.
Sanitization: There is no evidence of sanitization, escaping, or filtering applied to the content extracted from markdown files before it is processed or presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill's documentation and SKILL.md instruct the user to install well-known Python packages including chromadb and python-frontmatter to support database and metadata operations.

vector-db-launch