The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses package managers to install official instrumentation libraries.
Installs the standard traceloop-sdk (Python) and @traceloop/node-server-sdk (TypeScript) libraries via pip and npm.
Executes a local helper script tfy-api.sh to facilitate communication with the TrueFoundry API.
[PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface as it is designed to read and process application logs.
Ingestion points: Application logs are ingested via the tfy_logs_download tool call and shell-based API calls described in SKILL.md.
Boundary markers: There are no explicit delimiters defined for framing the log output in the agent's response.
Capability inventory: The agent has capabilities to execute shell commands, install software packages, and modify application source code for instrumentation purposes.
Sanitization: The skill provides a specific security warning to the agent, instructing it to review logs for sensitive information (PII, secrets, tokens) before presenting them or forwarding them to other services.
[SAFE]: The skill follows security best practices for secret management.
Uses placeholders for sensitive environment variables like TFY_API_KEY and TFY_BASE_URL.
Recommends using the platform skill's secret management capabilities instead of hardcoding credentials in the application code.

truefoundry-observability