Context

The skill is script-heavy and evolving fast. When Python dependencies are missing, agents currently discover failures late (mid-command) and then waste tokens installing packages ad-hoc. This also reduces reproducibility across machines.

Goal

Provide a one-shot, stdlib-only bootstrap script that:

• creates a local venv,
• installs the self-contained skill package (editable),
• optionally installs heavier indexing/embedding dependencies.

Non-Goals

• Do not implement a server runtime.
• Do not require users to maintain a separate

  requirements.txt

  .
• Do not install anything globally; keep installs within a repo-local venv.

Approach

• Add

  skills/kano-agent-backlog-skill/scripts/dev/install_prereqs.py

  (stdlib-only).
• Default behavior: create

  .venv/

  and install

  skills/kano-agent-backlog-skill

  with

  [dev]

  extras.
• Optional flag:

  --with-embeddings

  to install embedding/FAISS deps (best-effort).
• Document the entrypoint in repo

  README.md

  and

  AGENTS.md

  .

Alternatives

• Let scripts fail and ask the agent to install missing deps (current; wastes tokens).
• Commit a frozen lockfile and require exact sync (too heavy for pre-alpha demo).

Acceptance Criteria

• A single command exists that creates a venv and installs required deps for the skill scripts.
• The command is documented in

  README.md

  and

  AGENTS.md

  .
• The script uses only stdlib so it can run before any dependencies are installed.

Risks / Dependencies

• Some embedding deps are platform-dependent; keep them optional and best-effort.

Worklog

2026-01-10 02:05 [agent=codex] Planned: add a one-shot script to create venv and install required Python deps so agents don't waste tokens installing ad-hoc. 2026-01-10 02:11 [agent=codex] Ready: bootstrap script and docs are specified with acceptance criteria. 2026-01-10 02:11 [agent=codex] Done: added stdlib-only install script and documented usage in README.md and AGENTS.md.