菜单
Run credit-first local Agent Bridge orchestration in <PRIVATE_REPO_B> with strict read-only validation, bounded indexing, and fail-closed guidance hints. Excludes MCP Comfy diagnostics.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | repo-b-local-bridge-orchestrator |
| description | Run credit-first local Agent Bridge orchestration in <PRIVATE_REPO_B> with strict read-only validation, bounded indexing, and fail-closed guidance hints. Excludes MCP Comfy diagnostics. |
Use this skill to run a manual local-first Agent Bridge workflow in <PRIVATE_REPO_B> and emit validated guidance hints without cloud fallback.
/api/agent/* readiness/tasks, bounded file indexing, analyze_files validation, and guidance hint output./api/mcp/* operations and shim.comfy.* diagnostics.$repo-b-mcp-comfy-bridge.read_only, fail-closed)./health, /api/agent/capabilities).analyze_files task (allow_write=false, dry_run=true).guidance_hints JSON and stop on failures.Copy these files into <PRIVATE_REPO_B> before first run:
assets/repo_b/tools/local_bridge_orchestrator.py -> tools/local_bridge_orchestrator.pyassets/repo_b/tools/local_bridge_validate.py -> tools/local_bridge_validate.pyassets/repo_b/tools/local_bridge_orchestrator.example.env -> tools/local_bridge_orchestrator.example.envassets/repo_b/tests/tools/test_local_bridge_orchestrator.py -> tests/tools/test_local_bridge_orchestrator.pyassets/repo_b/tests/tools/test_local_bridge_validate.py -> tests/tools/test_local_bridge_validate.py$env:REPO_B_LOCAL_ORCH_ENABLED = "1"
$env:REPO_B_LOCAL_ORCH_CONFIDENCE_MIN = "0.85"
$env:REPO_B_LOCAL_ORCH_EVIDENCE_MIN = "2"
$env:REPO_B_LOCAL_ORCH_FAIL_CLOSED = "1"
$env:REPO_B_LOCAL_ORCH_MAX_HINTS = "12"
$env:REPO_B_CONTINUE_BRIDGE_ENABLED = "1"
$env:REPO_B_CONTINUE_MODE = "read_only"
$env:REPO_B_CONTINUE_ALLOWED_ROOTS = "G:\GitHub\<PRIVATE_REPO_B>"
Run from <PRIVATE_REPO_B> root:
python tools/local_bridge_orchestrator.py \
--task ticket-123 \
--prompt-file .codex/local_prompts/ticket-123.txt \
--scope connector \
--json-out .codex/local_bridge/ticket-123.json
Use your real usage history before scale-up:
python3 "$CODEX_HOME/skills/usage-watcher/scripts/usage_guard.py" analyze \
--input /path/to/usage.csv \
--window-days 30 \
--format table
python3 "$CODEX_HOME/skills/usage-watcher/scripts/usage_guard.py" plan \
--monthly-budget <credits> \
--reserve-percent 20 \
--work-days-per-week 5 \
--sessions-per-day 3 \
--burst-multiplier 1.5 \
--format table
Authority reminder:
http://127.0.0.1:9000/v1 and http://127.0.0.1:2337/v1 as authoritative model lanes.http://127.0.0.1:1234/v1 only as a non-authoritative operator surface.references/orchestrator-workflow.mdreferences/validation-contract.mdreferences/mx3-phase2-roadmap.mdIf this lane is unresolved, blocked, or ambiguous:
$skill-hub for chain recalculation.Run the local NullClaw host security app for skill governance, curated-source discovery, guarded threat suppression, and self-governance on Windows hosts.
Sweep one or more local repositories for implementation gaps and produce deterministic, evidence-backed remediation lanes. Use when you need cross-repo detection of missing tests, docs lockstep drift, risky TODO/FIXME additions, and release-hygiene misses.
Build and query bounded metadata-only repository indexes without rg process churn. Use when working in very large repos and you need deterministic file discovery by path, extension, language, scope, or freshness.
Detect recent cross-repo work and produce deterministic JSON evidence for skill curation and routing decisions. Use when workflows span multiple repos and you need a bounded view of recent commits, touched files, and policy/contract-sensitive changes.
Enforce hardware-first diagnosis and fixes in <PRIVATE_REPO_B>. Use for runtime probe, telemetry, inference, and integration failures where strict real-hardware behavior, no new stubs, deterministic diagnostics, and no unrequested driver/runtime mutation are required.
Run a defender-first security sweep on code, configs, prompts, model/tooling surfaces, or third-party contribution lanes. Use when a request involves safe bug, leak, zero-day-class, exploit, or hack hunting for protection, when contributing to outside repositories and you want a focused security pass, or when touching auth, secrets, permissions, network exposure, prompt/tool boundaries, data flow, or update/build surfaces. This skill is defensive only and must never be used for weaponization or unauthorized access.