Call agent
One skill that routes a task to the peer AI CLI best suited for it: Codex, Antigravity (agy), Kiro, Claude Code, or NotebookLM.
npx -y skills add cskwork/call-agent --skill call-agentAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Delegate one task to a named peer CLI, or when the host lacks the required capability.
SKILL.md
3.3 KB, 871 tokens by cl100k_base, as published. Nobody here has run it
call-agent - delegate to a peer AI CLI
Router. You (the host CLI) hand ONE task to a peer agentic CLI when the peer does it
better, then report the result back. Load only the one
reference/<target>/call.md you route to - never all of them.
Rule zero - never call yourself
Whichever CLI is reading this is the host. Routing a task to the host is a no-op (you
would just do it yourself). Delegate only to a DIFFERENT CLI whose capability the host
lacks. Inside Claude Code, never route to claude; inside Codex, never route to codex.
Trigger policy (conservative - external CLIs spend separate credits)
Fire on TWO conditions only:
- Explicit name - the user names a target tool.
- Capability gap - the user asks for something the host cannot do natively.
Never auto-fire on generic "review / plan / refactor / debug" - the host is presumed capable. When the win is unclear, ask before spending a peer's credits.
Route (match the request to ONE target, then load its call.md)
| Signal in the request | Target | Needs binary | Load |
|---|---|---|---|
"codex"; image generation; one-shot codex review | codex | codex | reference/codex/call.md |
| "agy" / "antigravity" / "gemini cli"; Google-grounded web search; image gen; science DB (gnomAD/UniProt/PubMed/ChEMBL) | agy | agy | reference/agy/call.md |
"kiro" / "kiro-cli"; natural-language -> shell translate; MCP cross-register; AWS Bedrock peer opinion | kiro | kiro-cli | reference/kiro/call.md |
| "claude" / "claude code"; implementation; 1M-context planning; deep review | claude | claude | reference/claude/call.md |
| "notebooklm" / "nblm"; RAG / audio / mind-map over a PDF/URL/YouTube corpus | notebooklm | notebooklm | reference/notebooklm/call.md |
| "gpt-pro" / "chatgpt pro" / "gpt5 pro"; deep-reasoning second opinion packaged for ChatGPT Pro (web, flat-fee subscription) | gpt-pro | none (tar/pbcopy) | reference/gpt-pro/call.md |
Tie-breakers: an explicit tool name always wins over a capability guess. Image
generation is offered by both codex (gpt-image-2, ChatGPT-auth, highest quality) and
agy - prefer codex when installed, fall back to agy.
Preflight (every route)
Each call.md defines its preflight. Run it before sending the task. Claude
implementation requires both auth and shell-capability preflights. If the host blocks
the target, report the normal-terminal fallback instead of retrying with weaker safety.
Reference map (load only what the routed target needs)
| Load | When |
|---|---|
reference/<target>/call.md | you routed to <target> (always) |
reference/agy/patterns.md, reference/codex/patterns.md | need orchestration recipes (parallel fan-out, async jobs) |
reference/<target>/reference.md | need the full flag table / gotchas (agy, codex, kiro, notebooklm) |
reference/agy/templates.md | need copy-paste agy prompt scaffolds |
reference/<target>/scripts/*.sh | the routed call.md tells you to run a wrapper script |
Done = target named; route-specific preflight passed; output and cost/session id reported; generated files verified. A blocked target ends with an actionable fallback, never a success claim.
What ships with it: 40 files
119.7 KB alongside SKILL.md, 27 of them executable
reference/
- agy/call.md2.5 KB
- agy/patterns.md2.4 KB
- agy/reference.md1.7 KB
- agy/templates.md1.6 KB
- agy/tests/smoke.shruns1.2 KB
- claude/call.md5.5 KB
- claude/scripts/claude-implement.shruns1.5 KB
- claude/scripts/claude-mcp-tools.shruns973 B
- claude/scripts/claude-plan.shruns1.3 KB
- claude/scripts/claude-review.shruns502 B
- claude/scripts/claude-review-stream.pyruns8.1 KB
- claude/scripts/preflight-auth.shruns1.1 KB
- claude/scripts/preflight-shell.shruns1.8 KB
- claude/tests/assert-live-mcp.pyruns1.2 KB
- claude/tests/smoke.shruns7.6 KB
- claude/tests/test-review-stream.pyruns4.8 KB
- codex/call.md4.7 KB
- codex/patterns.md4.4 KB
- codex/reference.md2.7 KB
- codex/scripts/codex-async.shruns10.0 KB
- codex/scripts/codex-imagegen.shruns1.2 KB
- codex/scripts/codex-review.shruns3.9 KB
- codex/tests/smoke.shruns4.5 KB
- gpt-pro/call.md4.4 KB
- gpt-pro/scripts/gpt-pro-bundle.shruns6.8 KB
- gpt-pro/scripts/gpt-pro-live.shruns6.2 KB
- gpt-pro/scripts/gpt-pro-preflight.shruns1.3 KB
- gpt-pro/tests/smoke.shruns2.1 KB
- kiro/call.md3.1 KB
- kiro/reference.md4.2 KB
- kiro/scripts/kiro-add-mcp.shruns1.0 KB
- kiro/scripts/kiro-chat.shruns921 B
- kiro/scripts/kiro-merge.shruns979 B
- kiro/scripts/kiro-preflight.shruns1.0 KB
- kiro/scripts/kiro-translate.shruns362 B
- kiro/tests/smoke.shruns2.1 KB
- notebooklm/call.md3.0 KB
- notebooklm/reference.md2.7 KB
- notebooklm/scripts/notebooklm-preflight.shruns2.0 KB
- notebooklm/tests/smoke.shruns2.5 KB