tenx-ai 0.1.3

Tenx assessment CLI

assessment-cli

CLI client for Tenx hiring assessments.

Highlights

• Uses the Claude Agent SDK for local interactive coding sessions.
• Sends cleaned chat messages to assessment-proxy over HTTP when configured.

Setup

python -m venv .venv
.venv\Scripts\activate
pip install -e ".[dev]"
copy .env.example .env

Set ANTHROPIC_API_KEY in .env before running.

Run

tenx-ai

The CLI starts a 60-minute assessment timer when the process starts. Use /time to display the remaining time.

Set TENX_ASSESSMENT_DURATION_SECONDS in .env to override the timer length.

Or:

python -m assessment_cli

Submit changes

Run this from your assessment workspace to stage, commit, and push:

tenx-submit

Optional commit message:

tenx-submit --message "Assessment submission"

After a successful push, tenx-submit starts an FRQ phase:

• It generates free-response questions from your most recent commit (HEAD).
• It prompts you for answers in the terminal.
• It saves the FRQ question/answer pairs to assessment-proxy.

Chat history persistence

Set these in .env to enable persistence:

• ASSESSMENT_PROXY_URL (for example http://localhost:8000)
• ASSESSMENT_PROXY_TOKEN (must match the proxy’s PROXY_BEARER_SECRET)

When ASSESSMENT_PROXY_URL is unset, the CLI runs without saving chat history to the proxy.

tenx-ai writes the active assessment session id to .tenx/assessment_session_id in your workspace so tenx-submit can reuse the same session for FRQ persistence.

Breaking change (Phase 0)

The proxy path and JSON shape changed: see assessment-proxy/README.md.

Security notes for Codespaces

• Do not store the Supabase secret API key in the CLI.
• Use only proxy URL and bearer token in CLI env.
• Treat ASSESSMENT_PROXY_TOKEN as sensitive and rotate if leaked.