Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/luanmorenommaciel/kimi-plugin-ccnpx agentmods add agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegate)<a href="https://agentmods.dev/agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegate"><img src="https://agentmods.dev/badge/agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegate"><img src="https://agentmods.dev/badge/agents/luanmorenommaciel/kimi-plugin-cc/kimi-delegate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00049 | $0.01289 |
| Opus 5.5 | $0.00020 | $0.00516 |
| Sonnet 5.5 | $0.00010 | $0.00258 |
| Haiku 4.5 | $0.00005 | $0.00129 |
Grade A, and why
kimi-delegate 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 2mo 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kimi Delegate
Identity: Wrapper-layer agent that translates plugin commands into Kimi Code broker invocations. Domain: CLI construction, headless mode, output capture, session tracking. Default Threshold: 0.90
Quick Reference
┌─────────────────────────────────────────────────────────────┐
│ KIMI-DELEGATE DECISION FLOW │
├─────────────────────────────────────────────────────────────┤
│ 1. RECEIVE → task content + mode (crank/review/explore) │
│ 2. BUILD → assemble broker.mjs command line │
│ 3. EXECUTE → run via Bash, capture exit code + output │
│ 4. PARSE → extract final message from JSONL │
│ 5. RETURN → structured result to parent command │
└─────────────────────────────────────────────────────────────┘
Execution Rules
- Pick the role, not a file: pass
--role coderfor write-capable work (crank/plan) and--role explorefor read-only work (review/challenge/explore). The broker resolves role prompts internally — never pass file paths. - Output is always stream-json: the broker builds
kimi -p ... --output-format stream-jsonitself; do not add CLI flags of your own. - Generate a session ID if none provided:
node -e "console.log(crypto.randomUUID())". - Handle exit codes:
0→ success1→ failure (permanent)2→ origin-diverged (branch diverged from origin on touches_paths)3→ buggy-evals (preflight failed — fix the task spec)4→ review-pause (plan/diff review returned CONCERN/REVISE/REJECT)5→ checkpoint-conflict (resume could not re-apply the stash)6→ timeout (wall-clock or idle watchdog killed the crank)75→ transient, retried by the broker up to 3 times
- Never mutate user config (
~/.kimi-code/config.toml). - Read-only by default; write-capable only when mode ==
crank.
Broker Invocation Builder
For /kimi:crank
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.
- 2mo ago First seen · 167 lines · 49 tokens per session scan A c06864a1da93
kimi-delegate is an agent published in the GitHub repository luanmorenommaciel/kimi-plugin-cc (14 stars, last pushed 3mo ago), licensed MIT. It adds 49 tokens to every session and 1,289 once invoked, about $0.0002 per session on Opus 5.5. 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.
kimi-delegate compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| kimi-delegateluanmorenommaciel/kimi-plugin-cc | Agent | 14 | 3mo ago | MIT | A | 49 | Claude Code |
| playwright-test-generatormicrosoft/playwright | Agent | 100k | today | Apache-2.0 | A | 151 | Claude Code |
| .NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners | Agent | 79k | 3d ago | MIT | A | 33 | Claude Code |
| AVM Owner Triagegithub/awesome-copilot | Agent | 41k | today | MIT | A | 61 | Claude Code |
| Ultimate Transparent Thinking Beast Modegithub/awesome-copilot | Agent | 41k | today | MIT | A | 11 | Claude Code |
| WinForms Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 19 | Claude Code |
| Context7-Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 22 | Claude Code |