Dev & Engineering local-developmentbootstrap-scriptpytestrufftmuxgit-worktreecli-testingmulti-agent

ClawTeam Local Development Skill

A repo-internal development skill for ClawTeam: standardize your local environment and validate multi-agent flows so every code change is verified quickly and reliably.

FollowSkills review · FSRS-2.0
Use with care
53/ 100 5-point scale 2.7 / 5
1 2 3 4 5 6
1Trust14 / 25 · 2.8/5

Script behaviors are disclosed in SKILL.md (writes ~/.bashrc, ~/.local/bin, fixed venv); link script errors out on non-symlink targets — decent data-flow disclosure. But bootstrap modifies ~/.bashrc and silently overwrites existing symlinks without explicit user confirmation, and no rollback/uninstall path is documented; deducted.

2Reliability10 / 20 · 2.5/5

Bootstrap gives a clear error when uv is missing and handles bashrc block idempotently; however, link script targets REPO_ROOT/skills/clawteam, an unverified path assumption inconsistent with the .agents/skills layout, key-path reproduction cannot be confirmed statically, and the test suite does not cover the skill's own scripts; deducted.

3Adaptability10 / 15 · 3.3/5

Audience and boundaries are explicit ('repository-development oriented, not a general end-user usage guide'), prerequisites (tmux, uv, CLI agents) listed, trigger semantics clear; but failure/non-fit boundaries are thin and no Chinese-environment notes; deducted.

4Convention9 / 15 · 3.0/5

Versioned (0.2.0), MIT license, well-layered docs; but SKILL.md version 0.2.0 conflicts with pyproject 0.3.0, no changelog or maintenance-ownership statement, hidden path assumptions; deducted.

5Effectiveness6 / 15 · 2.0/5

As a repo-internal dev bootstrap the goal is sound and the bootstrap/link/validate flow is directly usable; but static review cannot verify actual script output, and dependence on external CLIs (claude/codex) makes benefit claims hard to corroborate; deducted.

6Verifiability4 / 10 · 2.0/5

Repo has CI workflows and a tests/ directory (adapter tests cover the spawn layer), but CI and tests do not cover this skill's scripts; key claims (bootstrap works, link paths correct) rest on author statements without third-party reproduction; deducted.

Evidence confidence:Low Reviewed Sep 09, 2026 Reviewed revision 01198332ef92
Before you use it
  • The bootstrap script modifies ~/.bashrc and writes a launcher to ~/.local/bin; back up affected files before running — no official rollback steps are documented.
  • The link script links to REPO_ROOT/skills/clawteam; verify this path exists in the current checkout to avoid dangling symlinks.
  • SKILL.md version (0.2.0) conflicts with pyproject version (0.3.0); verify the actually installed version before use.
  • This is a static review; no scripts were executed, so the assessment does not constitute empirical verification of script behavior.
  • Smoke tests depend on tmux and external CLI agents (claude/codex); confirm reachability of related services from mainland-China networks.
See the full review method →

What does this skill do, and when should you use it?

clawteam-dev is the developer-facing skill bundled inside the ClawTeam repository, dedicated to changing, debugging, and validating ClawTeam itself. It ships two bootstrap scripts: one installs the repo into a fixed uv virtualenv and standardizes the clawteam command, the other symlinks the repo's skills into other local projects. It also prescribes a layered validation strategy (ruff plus targeted pytest) and full command sequences for end-to-end smoke tests with real CLI agents. Note that this is a repository-development skill, not a general end-user usage guide.

Runs scripts/bootstrap_clawteam_dev.sh: creates a fixed uv env at ~/.clawteam-venv, installs the current repo with dev dependencies, writes ~/.local/bin/clawteam, and adds a clawteam() shell function to ~/.bashrc. Runs scripts/link_local_clawteam_skills.sh /path/to/project: creates symlinks at ./.agents/skills/clawteam and ./.claude/skills/clawteam in a target project. Runs linting and tests: ruff check clawteam/ tests/ plus targeted or cross-module pytest runs. Executes local smoke tests: clawteam team spawn-team, task create, spawn, board show, task wait, and clawteam harness conduct for harness validation. Cleans up test teams with clawteam team cleanup.

  1. A ClawTeam contributor who just changed source code and needs the cheapest possible check that spawn, board, or task flows still work
  2. A developer reproducing a spawn / board / task / inbox / harness bug and debugging it against the current checkout
  3. A maintainer on a machine with multiple Python environments who wants the clawteam command to always point at one fixed environment
  4. A developer who wants another local project to reuse the repo's latest ClawTeam skills via symlinks instead of copies
  5. An engineer smoke-testing real multi-agent coordination end-to-end with a CLI agent (e.g. codex) in tmux or subprocess mode

What are this skill's strengths and limitations?

Pros
  • One-command bootstrap fixes a uv environment, eliminating clawteam command drift across Python environments
  • Explicit layered validation policy: ruff plus targeted pytest first, escalating to end-to-end only when needed, saving time
  • Complete, copyable smoke-test commands covering real spawn, task, board, and harness paths
  • Symlink script lets other local projects consume the repo's latest skills without reinstalling
Limitations
  • Narrow scope: serves the ClawTeam repository only; it does not teach end users how to use ClawTeam
  • Hard dependencies on tmux and the clawteam command; real-worker tests require a CLI agent that runs standalone
  • No evidence in the source about the quality of the test suite itself (coverage, CI status unmentioned)
  • Crashed tmux sessions or leftover worktrees must be inspected and cleaned up manually

How do you install this skill?

The skill lives at .agents/skills/clawteam-dev/ inside the ClawTeam repository and ships with it. Install the collection with pip install clawteam, or from source: git clone https://github.com/HKUDS/ClawTeam.git && cd ClawTeam && pip install -e . Prerequisites: the clawteam command available, tmux installed, and if you want real workers, a CLI agent such as claude or codex installed. The source does not document separate steps for copying this skill folder into other clients' skill directories.

How do you use this skill?

Inside the ClawTeam repo, run bash scripts/bootstrap_clawteam_dev.sh to set up the fixed environment, then source ~/.bashrc and confirm with clawteam --version. After code changes, run the smallest validation: ruff check clawteam/ tests/ and pytest tests/<target>.py -q; only escalate to broader suites (test_cli_commands.py, test_spawn_backends.py, test_harness.py) when changes cross modules. For real workflow validation, run the smoke sequence of team spawn-team, task create, spawn, board show, and task wait; for harness behavior use clawteam harness conduct dev-harness --goal ... --cli codex --agents 1. Finish with clawteam team cleanup.

How does this skill compare with similar options?

The README contrasts ClawTeam with other multi-agent frameworks (which require humans to write orchestration code and depend on Redis/message queues/Docker, while ClawTeam needs only a filesystem and tmux, with agents self-organizing via CLI commands); however, no comparable development-bootstrap tools are named for this skill itself, so no further comparison is drawn.

FAQ

How does this skill differ from the end-user skill in skills/clawteam/?
clawteam-dev is for developing, debugging, and validating the ClawTeam repository itself; the skills/clawteam/ entry mentioned in the README is the general skill that lets agents drive ClawTeam to accomplish tasks. Their scopes differ and should not be conflated.
What prerequisites are needed to run smoke tests?
You need the clawteam command available, tmux installed, and — for real workers — a CLI agent such as claude or codex that runs standalone on PATH. Your current directory must be the ClawTeam git repo when testing worktree isolation or local source changes.
When should I run broader tests?
The skill's guidance: prefer the smallest validation that proves the change (ruff plus a single pytest file); only escalate to wider suites like test_cli_commands, test_spawn_backends, or test_harness — and harness-level smoke tests — when the change crosses spawn/runtime/workflow boundaries.
How do I clean up after a failed or crashed test run?
Prefer the real commands: clawteam team cleanup dev-smoke / dev-harness. If a crashed tmux session or worktree is left behind, inspect it first and then clean it up explicitly rather than editing state files directly.

More skills from this repository

All from HKUDS/ClawTeam

Related skills