The Agent Skills Directory

[PROMPT_INJECTION]: The SKILL.md file contains a template for agent invocation that includes mode: "bypassPermissions". This is a direct instruction to bypass platform-level safety filters and permission constraints.
[PROMPT_INJECTION]: The skill ingests untrusted data from transcripts and manuals without defining security boundaries, leaving it vulnerable to indirect prompt injection.
Ingestion points: The skill reads manual.md and transcript.txt files from the local filesystem to extract knowledge.
Boundary markers: None. There are no instructions or delimiters used to separate the source text from the agent's instructions.
Capability inventory: The skill has the capability to read files from the filesystem and write structured JSON output.
Sanitization: The skill does not implement any validation or sanitization of the processed text to prevent embedded instructions from being executed.

concept-extractor