donaldfilimon/abi / .claude/skills/mcp-smoke
mcp-smoke
VerifiedBuild the abi MCP server and smoke-test its JSON-RPC tool surface — start `abi-mcp`, send a `tools/list` over the stdio transport, and assert the frozen 12-tool contract (ai_run, ai_complete, ai_learn, ai_train, wdbx_query, scheduler_stats, scheduler_info, connector_test, gpu_status, plugin_list, wdbx_stats, plugin_run). Use to run/start/smoke-test abi-mcp, verify the MCP tool list, or check the 12-tool contract still holds. Local stdio; the process exits on pipe EOF.
mcp-smoke — assert the MCP server's 12-tool contract
Driver: .agents/skills/mcp-smoke/smoke.sh (paths relative to repo root).
Builds release abi-mcp via ./tools/cargo.sh, launches it through
mcp/launcher.sh stdio, sends one JSON-RPC tools/list over
stdio, and asserts the frozen 12-tool set. Fully local — no network.
The driver overrides inherited store settings with ABI_WDBX_PATH=:memory:
and ABI_WDBX_PERSIST=0. Locked builds explicitly use this checkout's target/
directory, even with an inherited CARGO_TARGET_DIR. Python 3 parses the
response and checks the JSON-RPC version, request ID, success envelope and
tool names. A failed build or server exit fails the smoke.
Run (agent path)
.agents/skills/mcp-smoke/smoke.sh
Prints RESULT: PASS — 12/12 frozen MCP tools present (exit 0) or RESULT: FAIL.
One-liner:
printf '%s\n' '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' \
| ABI_WDBX_PATH=:memory: ABI_WDBX_PERSIST=0 ./mcp/launcher.sh stdio \
| jq -r '.result.tools[].name'
Gotchas
- ⚠️ Do NOT count tools with
grep -c '"name":'— it returns 13, not 12. Usejq -r '.result.tools[].name'or count only tool-list name values. - stdio framing is newline-delimited JSON — not LSP
Content-Length. - No hang, no
timeoutneeded. Pipe EOF exits the stdio loop. - Optional loopback HTTP/SSE may log on stderr; drop stderr when grepping stdout.
Troubleshooting
| Symptom | Fix |
|---|---|
build FAIL | ./tools/cargo.sh build --locked --release -p abi-mcp then ./tools/check.sh |
| empty response | trailing \n required; drop stderr |
| got 13 tools | over-counting schema "name" fields |
| tool set mismatch | reconcile crates/abi-mcp/src/handlers.rs with AGENTS.md frozen list |
Historical golden: tests/golden/mcp-tools-list.json.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |