The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs the agent to access and analyze sensitive file paths, specifically .env files, to identify environment variables and database connection strings as part of its 'Survey Phase' and 'Dependency Mapping' processes (found in references/patterns.md). While this is intended for context gathering, it creates a surface for sensitive data exposure.
[COMMAND_EXECUTION]: The skill provides specific command-line templates for git (e.g., git shortlog, git log, git blame) to be used for history reconstruction and identifying code authorship (found in references/patterns.md).
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data in the form of legacy codebases and their associated metadata.
Ingestion points: The agent reads source code, README files, deployment configurations, database schemas, and test files (references/patterns.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the analyzed code are provided.
Capability inventory: The agent has the capability to read local files and potentially execute shell commands (git) to perform its task.
Sanitization: No sanitization or validation of the ingested code content is mentioned.

legacy-archaeology