The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to determine the installation path and git context.
SKILL.md: Resolves ${PLUGIN_ROOT} using a shell script: PLUGIN_ROOT="${CLAUDE_PLUGIN_ROOT:-}"; [ -n "$PLUGIN_ROOT" ] && PLUGIN_ROOT="$(cd "$PLUGIN_ROOT" 2>/dev/null && pwd)"; echo "PLUGIN_ROOT=$PLUGIN_ROOT".
references/maintenance-tasks.md and references/user-story-tasks.md: Executes git branch --show-current to check the active branch and git config user.name to identify the assigned user.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources within the project repository to generate planning documents.
Ingestion points: Reads content from docs/specs/user-stories/*/README.md, docs/specs/work-items/*/README.md, and dossiers from the work-research skill.
Boundary markers: The skill does not define explicit boundary markers or delimiters when parsing these files, though it looks for specific Markdown sections.
Capability inventory: The agent can execute shell commands, write new specification files to the disk, and invoke other functional skills like work-implement or test-define.
Sanitization: There is no evidence of sanitization or filtering applied to the ingested content before it is processed or used to influence downstream skill invocations.

work-plan