Complete AgentTeams test cycle including installation, uninstallation, running tests, and exporting debug logs for analysis. Use for (1) verifying AgentTeams functionality (2) CI/CD test validation (3) issue diagnosis and debugging (4) pre-merge testing. Trigger words: test AgentTeams, run AgentTeams tests, agentteams test, make test, verify AgentTeams installation.
90%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Detected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill is a legitimate AgentTeams test workflow, but `run-agentteams-test.sh` automatically runs `git reset --hard origin/main` on any existing repo it detects, which can silently discard the user's uncommitted local changes. That is a careless destructive step, not evidence of malicious intent.
The skill prompts the agent to compromise the security or integrity of the user’s machine by modifying system-level services or configurations, such as obtaining elevated privileges, altering startup scripts, or changing system-wide settings.
The skill has the agent run `make install` with `AGENTTEAMS_YOLO=1`, which skips confirmations, and `make uninstall`, which changes machine state. It never asks for sudo, creates user accounts, or edits system or SSH/systemd files, and the install targets a project the user is testing, so the risk is moderate rather than clear.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required `analyze` path exports Matrix room messages via `export-debug-log.py` and the agent reads their `body` previews from any sender in the time range, so free text from chat rooms the workflow consumes reaches the agent without it selecting a specific item.
agentscope-ai/AgentTeams
bc2c064
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.