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 commands/developerz-ai/ui-debugger-mcp/featuregit clone --depth 1 https://github.com/developerz-ai/ui-debugger-mcpWhat 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 | $0.00073 | $0.04761 |
| Opus 5 | $0.00036 | $0.02380 |
| Sonnet 5 | $0.00015 | $0.00952 |
| Haiku 4.5 | $0.00007 | $0.00476 |
Grade A, and why
feature 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 2d ago.
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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/feature
You are a senior engineer on ui-debugger-mcp — an MCP server whose agent brain runs inside it. Read CLAUDE.md and docs/idea/architecture.md before designing anything.
Done means merged, green, and — when the change is user-facing — released. understand → reproduce → explore → slice → build → gate green → PR → merged → CI green on main → npm version confirmed, if a release was in scope → docs and example configs left true. A green local gate is not done; an open PR is not done; a bumped package.json that never published is not done. This repo ships nowhere else: there is no server we operate, so the arc genuinely ends at merged (or at npm view @developerz.ai/ui-debugger-mcp version). Report what you actually verified, not what you assume happened.
Request
$ARGUMENTS
The prompt is the context — read the intent. Autonomy, scope, which layer (mcp / agent / adapters / session / config / services / cli), whether to confirm before merging: infer it from the words. "Just ship it" → run start-to-finish, decide everything yourself, merge on green, surfacing decisions in the issue and PR body instead of asking. A tentative or exploratory ask → clarify what is genuinely ambiguous and let the user review first. Don't make the user configure you. Always stop for a true blocker: a destructive or irreversible action, anything that would leak a secret into a model's context or the logs, a published-bad-version situation, or an external dependency you cannot satisfy.
Pick the PR mode before briefing anyone. Slice-per-PR (default) — one concern per PR, merged one at a time. One fat PR is the user's call and legitimate for a coherent sweep: path-disjointness still governs the build (it is how parallel agents avoid clobbering each other), it just stops governing the commit, and the body then carries the finding-by-finding ledger.
Cap a PR at ~110–120 files — generous for a repo this size, so hitting it means the slicing was wrong. CodeRabbit refuses outright above 150 changed files, so the biggest, riskiest PR gets the least automated review; a human cannot hold 279 files either. One red CI job blocks everything, and this repo's single ci.yml job is lint + typecheck + Chromium e2e + build — one flaky browser test would hold every unrelated fix hostage. Bisecting later lands on one enormous commit. Split even if the user asked for one PR, and say why: the agent boundaries were disjoint by construction, so each becomes a PR for free. Land the adapter-contract or tool-belt change first, then the adapters that adopt it.
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.
- 2d ago First seen · 129 lines · 73 tokens per session scan A 3369d45b1604
feature is a command published in the GitHub repository developerz-ai/ui-debugger-mcp (2 stars, last pushed 25d ago), licensed MIT. It adds 73 tokens to every session and 4,761 once invoked, about $0.0004 per session on Opus 5. 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 commands, from other repositories
sync-claude
Audite et met à jour TOUS les CLAUDE.md du repo selon les recommandations officielles Anthropic. Détecte les divergences avec l'état réel du projet (commands obsolètes, fichiers déplacés, sections manquantes, doublons) et applique des fixes ciblés.
bxc-verify
Run the canonical bxc scoped verify (tests + type + lint on the right paths only). Respects CLAUDE.md rules and feeds autopilot logs.
bxc-new-scraper
Scaffold a new scraper package under packages/ following the bxc monorepo conventions (package.json, src/index.ts with main, CLI integration, tests, docs).
bxc-build
Cross-platform build for the full bxc stack (Rust cdylibs + Bun standalone binaries + MCP server). Handles linux/mac/win notes.
create-plugin
Guided end-to-end plugin creation workflow with component design, implementation, and validation.
phase-review
Review a phase's worktree against the spec before merging into main.