The Agent Skills Directory

[EXTERNAL_DOWNLOADS] (HIGH): The skill clones two external repositories (timecamp-cli and good-enough-timecamp-data-pipeline) from github.com/timecamp-org. This organization is not on the pre-approved Trusted External Sources list.
[REMOTE_CODE_EXECUTION] (HIGH): The skill executes the downloaded code using npm link and uv run --with-requirements requirements.txt. This allows for arbitrary code execution from an untrusted external source.
[COMMAND_EXECUTION] (MEDIUM): The skill frequently uses subprocess calls and shell commands to manage repositories, install dependencies, and run analytics pipelines.
[INDIRECT_PROMPT_INJECTION] (HIGH):
Ingestion points: The skill ingests untrusted data from the TimeCamp API (entries, tasks, computer activities) via dlt_fetch_timecamp.py and processes it using DuckDB (SKILL.md).
Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the processing logic.
Capability inventory: The skill possesses high-privilege capabilities including shell command execution (git, npm, uv, duckdb) and file system access.
Sanitization: There is no evidence of sanitization or filtering of the external TimeCamp content (such as entry notes or task names) before it is processed by the agent or the database.

timecamp