The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands including git, ls, grep, and sed to retrieve the repository root, current branch name, and session metadata from internal transcript logs. These operations are restricted to local information gathering necessary for generating the report.
[EXTERNAL_DOWNLOADS]: The documentation references installation via npx and a GitHub repository under the pilniczek organization. These are legitimate vendor-provided resources for skill distribution and updates.
[INDIRECT_PROMPT_INJECTION]: The skill incorporates content from existing WORK-REPORT.md files located in the user's working directory to maintain continuity across sessions.
Ingestion points: Reads existing WORK-REPORT.md in Step 2 of SKILL.md to carry forward abandoned items and follow-ups.
Boundary markers: No explicit delimiters are used to wrap content imported from the existing file.
Capability inventory: The skill is authorized to use Bash (git and unix utilities), Read, Write, and Edit tools.
Sanitization: Step 4 of SKILL.md explicitly directs the agent to redact secrets like API keys, passwords, and PII, and to reference artifacts rather than reproducing their content.

work-report