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 skills add Negai-ai/AgentClaw --skill agent_creatorgit clone --depth 1 https://github.com/Negai-ai/AgentClawWrote 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/skills/negai-ai/agentclaw/agent_creator)<a href="https://agentmods.dev/skills/negai-ai/agentclaw/agent_creator"><img src="https://agentmods.dev/badge/skills/negai-ai/agentclaw/agent_creator.svg" alt="Measured on agentmods" 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 | $0.00048 | $0.06171 |
| Opus 5 | $0.00024 | $0.03086 |
| Sonnet 5 | $0.00010 | $0.01234 |
| Haiku 4.5 | $0.00005 | $0.00617 |
Grade B, and why
agent_creator scanned grade B with 2 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 8d 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
- `shell(command="curl -s -N -X POST {BASE_URL}/_internal/api/workflow/run -H 'Content-Type: application/json' -d '{\"workflow_id\": \"my_workflow\", \"inputs\": {...}, \"response_mode\": \"streaming\"}'", timeout=300)` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `shell(command="curl -s -N -X POST {BASE_URL}/_internal/api/workflow/run -H 'Content-Type: application/json' -d '{\"workflow_id\": \"my_workflow\", \"inputs\": {...}, \"response_mode\": \"streaming\"}'", timeout=300)` How it starts
The opening of the file, as written. The whole thing — 551 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Creator (Lean Playbook)
This file is the default workflow-building guide.
Read this file first. Route into references/* when the task pattern calls for it.
0) Default Mode (Important)
Use Lean Mode by default:
- Keep to the shortest executable path.
- Prefer local patch fixes over full-file rewrites.
- Validate after each Python edit.
- Prefer not to continue to the next gate while the current gate is failing.
- Report
completed/partial/blockedtruthfully.
Keep the build lean, but do not skip a pattern reference when the request clearly matches it. The right reference is part of the shortest reliable path.
0.1) Reference Routing (Pattern Library)
Use references as early design aids, not last-resort repair manuals.
Before designing or coding, scan the user request for domain patterns:
| If the request mentions... | Read before design | Use it to choose... |
|---|---|---|
| SQL, NL2SQL, database Q&A, table, column, schema, analytics, dashboard data, log audit, compliance/audit report, scheduled data report | references/nl2sql.md |
tool-based exploration vs. process-based workflow, schema discovery, SQL validation, execution/report gates |
If no pattern matches, stay in this file and load references only when a concrete missing detail blocks progress.
0.2) Agent Construction Mental Model
A good AgentClaw workflow is not just code that runs; it is a small evidence pipeline that turns user intent into safe action.
Design the workflow in layers:
- Input contract: define what the user supplies and what output they expect.
- Evidence/discovery: gather facts the agent should not guess: files, schemas, APIs, current state, configs, and user rules.
- Reasoning/generation: use
LLMNodefor interpretation, synthesis, SQL/code/report drafting, natural-language explanation, and ambiguous mapping. - Validation/gating: use deterministic checks before execution, mutation, file writing, scheduling, or final claims.
- Action/output: execute only validated actions; write files or call APIs in small deterministic nodes.
- Final response: tell the user what happened, what was verified, and what remains blocked.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 551 lines · 48 tokens per session scan B 8f8f0218a748
agent_creator is a skill published in the GitHub repository Negai-ai/AgentClaw (341 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 48 tokens to every session and 6,171 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…