Skip to content

mikeng-io/agent-skills / skills/bridge-claude

bridge-claude

Verified

Reference adapter for Claude (Anthropic) dispatch. Read by any orchestrating skill via the Read tool. Defines how to invoke Claude sub-agents or the Anthropic API, with availability checks. Usable by any AI orchestrator — Claude Code, OpenCode, Codex, Gemini, or custom agents.

FileKindSize
README.mdmarkdown1.7 KB
SKILL.mdskill md14.2 KB
cli-reference.mdmarkdown3.2 KB
# Bridge: Claude

Reference adapter for Claude Code sub-agent dispatch. Part of the Deep Skills Suite bridge layer.

## What is this?

This is a **reference document**, not a runnable skill. Any orchestrating skill reads `SKILL.md` via the `Read` tool and embeds its instructions into Task agent prompts. The bridge defines how to spawn Claude sub-agents as internal reviewers.

## How it works

1. An orchestrating skill reads `bridge-claude/SKILL.md`
2. Spawns a Task agent (the "bridge executor") with these instructions embedded
3. Bridge executor spawns parallel domain expert sub-agents + DA + Integration Checker
4. Bridge executor collects findings and returns a `bridge_claude_report`
5. The calling skill receives the report for synthesis

## Why context: reference?

Claude bridges work via the Task tool — they spawn sub-agents internally. There's no CLI to invoke. The bridge is instructions, not a skill that needs to be "called" separately.

## Availability

Conditional — depends on what's available to the executor:

1. **Task tool** — when Claude Code is the executor (native sub-agent dispatch)
2. **Claude Code CLI** (`which claude`) — when any other executor can shell out to `claude -p`
3. **Anthropic API** — `ANTHROPIC_API_KEY` in environment
4. **SKIPPED** — none of the above available

Bridge-claude is non-blocking — it returns SKIPPED if no Anthropic access is available.

## Output

Returns a `bridge_claude_report` JSON with findings, verdict, and confidence level.

## Part of

- Deep Skills Suite
- Consumed by: any orchestrating skill (e.g., `deep-council`, `deep-review`, `deep-audit`, or custom skills)
- Depends on: `domain-registry` (for expert role selection)
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall