The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs several shell operations to manage the repository state.
It executes Git commands including git fetch, git pull --rebase, and git push to synchronize changes with the remote repository.
It attempts to execute a local Python script python3 scripts/update_readme_ideas_index.py if the project is identified as an 'ideas repo'.
[DATA_EXPOSURE]: The skill reads from specific configuration files in the user's home directory to determine project paths.
Evidence: Accesses ~/.config/ideas-root.txt and ~/.openclaw/ideas-root.txt during the input resolution phase in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data which could contain malicious instructions.
Ingestion points: Reads content from prd.md, tad.md, ux_design.md, and brand_kit.md (documented in agents/requirements-extractor.md).
Boundary markers: The instructions do not specify explicit delimiters or 'ignore' instructions for the sub-agents when parsing these files.
Capability inventory: The agent can write files (tasks.md), execute local Python scripts, and push changes to a Git repository (SKILL.md).
Sanitization: No explicit sanitization or validation of the input markdown content is performed before processing.

tasks-generator