The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a git pull command to synchronize data within the identified CEOS repository root.
[EXTERNAL_DOWNLOADS]: Synchronization involves fetching updates from the repository's configured remote origin during the sync process.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and summarizes data from multiple local files to provide context for the checkup.
Ingestion points: Reads content from data/vision.md, data/accountability.md, data/rocks/, data/scorecard/, data/people/, and data/issues/.
Boundary markers: No specific delimiters or instructions are provided to the agent to treat ingested file content as untrusted data or to ignore embedded instructions.
Capability inventory: The skill has permissions to write to the data/checkups/ directory and execute shell commands like git pull.
Sanitization: There is no logic specified to sanitize or validate the data read from context files before it is presented or used in the checkup logic.

ceos-checkup