Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/debugswift/xcodebazelmcp/testinggit clone --depth 1 https://github.com/DebugSwift/XcodeBazelMCPWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00548 |
| Opus 5 | $0.00000 | $0.00274 |
| Sonnet 5 | $0.00000 | $0.00110 |
| Haiku 4.5 | $0.00000 | $0.00055 |
Grade A, and why
testing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 31 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Testing
Framework
230 unit tests via Vitest, coverage via @vitest/coverage-v8.
Test Organization
| File | Covers |
|---|---|
bazel-tools.test.ts |
Tool name set (order-independent via .sort()), startupArgs schema, required fields, streaming flag, unknown tool rejection |
handlers/handlers.test.ts |
Every handler exports non-empty definitions, canHandle correctness, handle returns undefined for unknown tools, no duplicates across handlers, total tool count (112) |
helpers.test.ts |
stringOrUndefined, numberOrUndefined, prependWarning, applyDefaults (default merging vs explicit arg priority) |
simulators.test.ts |
findAppBundle, readBundleId — uses real filesystem fixtures (.test-fixtures/ dir, created in beforeAll, cleaned in afterAll) |
bazel.test.ts |
Arg builders, label validation, query sanitization, YAML config parsing (flat + nested profiles, enabledWorkflows, boolean/numeric values, comments, empty content, activateProfile errors) |
workspace.test.ts |
assertBazelWorkspace (non-existent/file/missing markers/valid), readBspStatus (missing dir, valid JSON, malformed JSON) |
upgrade.test.ts |
upgradeHint fallback, compareVersions edge cases, detectInstallMethod |
| MCP smoke test | Server initializes without error |
Key Conventions
- All tests are pure unit tests — no actual Bazel builds or simulator boots in CI.
- When refactoring handlers into separate modules, tool name ordering in tests must use
.sort()since import order is not guaranteed. - Many tool handlers are hard to unit-test (they shell out to
bazel,simctl,devicectl). E2E testing againstexample_projects/BazelAppis the primary validation. - MCP JSON-RPC can be tested by piping
printfJSON lines intonode dist/cli.js mcp. Usesleepbetween sequential calls since the server processes sequentially.
Known Test Infrastructure Issues
- Global mutable
config+configLoadedinruntime/config.tsleaks state across test files. No current tests callrunBazelorassertBazelWorkspaceso it doesn't cause failures — watch out when adding integration tests. - Module-level mutable counters (
logCaptureCounter,videoRecordingCounter) inhelpers.tshave no reset function. Tests share state across runs.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 31 lines · 0 tokens per session scan A 7b2342c1bb2e
testing is an agent published in the GitHub repository DebugSwift/XcodeBazelMCP (7 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 548 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
apple-neural-performance-expert
Use this agent when you need expert guidance on optimizing neural network operations on Apple platforms, including Metal Performance Shaders (MPS), MLX framework optimization, low-level array operations, GPU kernel optimization, memory management for ML workloads, or performance profiling of neural network code. This…
code-search-ast-grep
Use this agent when you need to search for specific code patterns, syntax structures, or text across files in the current directory. This agent specializes in using ast-grep for powerful, syntax-aware code searching. Examples:\n\n \nContext: An agent needs to find all function definitions that match a certain…
task-agents
The primary agent workflow — inference setup resolution, the automation switch, evidence-first execution, tool policy, and the proposal/confirmation loop.
persistence-and-sync
The agent.sqlite entity and link model, bulk-read chunking, and exactly what leaves the device.
ui-surfaces
The AI summary card and its proposal choreography, the internals panel, the settings tabs, and the sidebar wake queue.
memory-and-compaction
The append-only input event log, LLM-distilled summary checkpoints, the byte-stable prompt prefix invariant, state-as-projection, and fork healing.