The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several standard shell commands such as fd, find, rg, grep, wc, and git to discover files and search for patterns within the target codebase. While these are necessary for the skill's functionality as a reconnaissance agent, they represent a significant capability surface for interacting with the local filesystem.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the codebase being scanned.
Ingestion points: The agent is instructed to read various source code files (e.g., .ts, .js, .py, .go) and follow imports to map the architecture.
Boundary markers: The instructions do not define any delimiters or specific prompts to help the agent distinguish between its instructions and the content of the files it is reading.
Capability inventory: The agent has the ability to execute shell commands and run local Node.js scripts (doc-lookup.cjs, context7-api.cjs) based on its findings.
Sanitization: There are no instructions provided to sanitize or validate the content extracted from the source files before the agent processes them.

recon