The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a Bash tool to execute a series of shell commands for detecting feature changes, finding documentation files, and extracting feature names from source code. This includes usage of git diff, find, grep, and sed to automate the audit process.
[EXTERNAL_DOWNLOADS]: During the validation phase, the skill runs npx markdownlint and npx markdown-link-check. These commands trigger the download and execution of utility packages from the official NPM registry, which is a well-known service for developer tooling.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted data from source code and documentation files.
Ingestion points: The skill reads .ts, .tsx, and .md files in the repository using grep and sed to extract feature labels.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore any potential command-like text embedded within the extracted feature names.
Capability inventory: The skill possesses Bash, Write, Edit, and Read capabilities, and it can invoke other skills like documentation-audit.
Sanitization: The extraction logic (sed 's/.*feature:\s*["'\']\([^"'\']*\)["'\'].*/\1/') performs simple regex extraction but does not sanitize the resulting strings for potential malicious instructions before including them in reports or UI prompts.

features-documentation