메뉴
This skill should be used when the user asks to spawn or generate an ARENA lab from a spec, run a spec through the pipeline, run several specs in parallel, bring up / start up a generated lab's Docker container for inspection, or find and run a finished run's image.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
This skill should be used when the user asks to query Logfire logs, inspect Logfire traces, fetch SQL trace data, get spans for a trace, inspect LLM spans, inspect tool calls, or check token usage from Logfire. Also use it to triage an ARENA spawn run — diagnose why a run failed, timed out, or took too long, get per-stage turns/cost/tokens, or compare a prod trace to a local one.
Add a single supporting (non-vulnerable) feature to a working ARENA skeleton app — a catalog, detail page, profile, order history, cart, etc. Triggered when the prompt contains "add-supporting-feature", "ARENA supporting feature", or a Feature block whose vuln is "none". Builds a real, reachable, working page wired into the shared nav. No vulnerability, no solution writeup.
Add a single vulnerable feature to a working ARENA skeleton app. Triggered when the prompt contains "add-vuln-feature", "ARENA feature writer", or a Feature block. Implements the feature and writes a human-readable solution writeup. A usability judge then browser-drives the feature and an exploit verifier independently reproduces the vulnerability against the running lab.
Styling for the build-capable SPA stacks (react-express, vue-fastapi): Tailwind CSS (+ shadcn/ui on React) themed per-run through CSS-variable design tokens. Invoke on these stacks to give each lab a distinctive, coherent, believable-product look that is different every run.
Styling for the server-rendered / no-build stacks (php, vanilla-flask): a token-driven CSS sheet themed per-run by editing one `:root` variable block. Invoke on these stacks to give each lab a distinctive, coherent, believable look — different every run — without a build step.
Turn an ARENA plan into per-feature implementation contracts (feature_contracts.json) so feature authors can run in parallel without colliding. Triggered by "ARENA feature planner" / "plan-features". Writes feature_contracts.json only — no code or builds.
| name | run-arena-lab |
| description | This skill should be used when the user asks to spawn or generate an ARENA lab from a spec, run a spec through the pipeline, run several specs in parallel, bring up / start up a generated lab's Docker container for inspection, or find and run a finished run's image. |
| version | 1.0.0 |
Spawn a vulnerable lab from a spec and bring its container up for inspection. All
commands run from the spawner/ directory.
cd spawner
uv sync # once, to set up the venv
uv run arena specs/leaky-query.yaml
arena = "arena.cli:main" (spawner/pyproject.toml); a bare spec arg
routes to the spawn command (spawner/arena/cli.py).spawner/specs/*.yaml. Ad-hoc/showcase specs are conventionally
kept in .context/showcase-specs/ (gitignored) so they don't clutter the repo.[arena] .spawner/arena/cli.py:83-88)| Marker | Meaning |
|---|---|
[arena] SUCCESS followed by the manifest JSON | run finished; image built |
[arena] FAILED — reason=<reason> | run failed at a stage |
[arena] workdir preserved at: <path> | inspect this dir to debug a failure |
Failure reasons come from spawner/arena/spawn/errors.py: plan_failed,
skeleton_failed, feature_failed, supporting_failed, usability_failed,
sanity_failed, plus max_turns, wallclock, manifest_missing, image_missing,
goals_mismatch. For why a stage failed, hand off to the logfire skill
(trace triage).
spawner/arena/config.py)| Var | Default | Effect |
|---|---|---|
ARENA_MODEL | claude-sonnet-4-6 | planner + sanity model |
ARENA_IMPL_MODEL | claude-opus-4-8 | scaffold + feature model |
ARENA_USABILITY_MODEL | = ARENA_MODEL | usability browser judge |
ARENA_<STAGE>_MAX_TURNS | per-stage | turn cap (e.g. ARENA_SKELETON_MAX_TURNS=50) |
ARENA_<STAGE>_WALLCLOCK_SECONDS | per-stage | wall-clock cap |
ARENA_RUNS_DIR | ~/.arena/runs | where run workdirs land |
ARENA_DOCKER_URI | arena-local | image tag prefix (a dotted host → push to registry) |
ARENA_SKELETON_MAX_TURNS=50 ARENA_IMPL_MODEL=claude-opus-4-8 uv run arena specs/leaky-query.yaml
A run writes to ~/.arena/runs/<spec.id>-<8hex>/ (config.py:38,
spawn/runroot.py). The agent's output is under work/:
~/.arena/runs/<spec.id>-<8hex>/work/
app/ generated app (Dockerfile + source + db init)
exploits/ one runnable exploit per vuln feature
solutions/ human-readable write-ups
plan.json planner output
manifest.json final descriptor (written by sanity)
The workdir is preserved on failure, so a FAILED run is still inspectable.
There is no built-in batching — run each spawn as its own background task and poll its log (the pattern used across the edinburgh/tashkent sessions):
cd spawner
mkdir -p ../.context/spawn-logs
for spec in react-shop vue-magazine vanilla-ops php-bank; do
uv run arena "../.context/showcase-specs/$spec.yaml" \
> "../.context/spawn-logs/$spec.log" 2>&1 &
done
wait
grep -lE 'SUCCESS|FAILED' ../.context/spawn-logs/*.log # which finished, and how
Prefer the harness's background-task mechanism (it re-invokes you on completion)
over a raw & when driving this interactively.
The final image is tagged arena-local:<spec.id>-<8hex> (spawn/image.py:13,
built by verify/docker.py). To run it:
docker images 'arena-local:*' # find the tag
cid=$(docker run -d --rm -P arena-local:<tag>) # -P publishes exposed ports
docker port "$cid" 5000 # spec.port defaults to 5000
curl -fsS http://127.0.0.1:<host_port>/api/health # health endpoint (verify/docker.py:147)
/api/health returning any non-5xx (incl. 401/403) means the app booted. Seeded lab
credentials (e.g. admin@ares.local / ArenaAdminPass!1) are deliberately public
lab creds, not secrets.
Note:
agent-browser: command not found(exit 127) inside a run is a runtime packaging concern for the agent, not a problem with bringing up the container.
For local vs prod generation and image destinations, see
agent_docs/environments.md.