The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection within its research workflow. It explicitly instructs the agent to fetch and extract content from the web using search tools or text-based browsers.
Ingestion points: The Research stage in miniproject/SKILL.md uses brave_search and the lynx CLI to gather information from external websites.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands when processing the content fetched from the web.
Capability inventory: Across its components, the skill has the ability to write to the filesystem, execute shell scripts in the scripts/ directory, and perform git operations.
Sanitization: The instructions do not mention any sanitization, validation, or filtering of the content retrieved from external sources before it is incorporated into the agent's context.
[COMMAND_EXECUTION]: The skill includes several functional scripts to manage project identifiers, migrate data structures, and validate file schemas.
Evidence: miniproject/scripts/get-memory-dir.sh, miniproject/scripts/migrate-phases-to-inline.sh, miniproject/scripts/schema.ts, and storage-zk/scripts/wiki.
Hardcoded Path: The miniproject/SKILL.md file contains a hardcoded absolute path (/home/zenobius/.pi/agent/skills/projectmanagement/miniproject/scripts/get-memory-dir.sh) that points to the author's local environment, which may cause execution failures for other users.
[EXTERNAL_DOWNLOADS]: The skill integrates with external tools by recommending their use via package runners.
Evidence: inmemoria/SKILL.md suggests running npx in-memoria to initialize and maintain codebase intelligence databases.

projectmanagement