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/OpenDCAI/Dataflow-LoopAInpx agentmods add skills/opendcai/dataflow-loopai/obtainerWrote 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/opendcai/dataflow-loopai/obtainer)<a href="https://agentmods.dev/skills/opendcai/dataflow-loopai/obtainer"><img src="https://agentmods.dev/badge/skills/opendcai/dataflow-loopai/obtainer/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/skills/opendcai/dataflow-loopai/obtainer"><img src="https://agentmods.dev/badge/skills/opendcai/dataflow-loopai/obtainer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 107 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
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 | $0.00190 | $0.09889 |
| Opus 5.5 | $0.00076 | $0.03956 |
| Sonnet 5.5 | $0.00038 | $0.01978 |
| Haiku 4.5 | $0.00019 | $0.00989 |
Grade A, and why
obtainer 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 1mo 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 — 766 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Obtainer Skill
Purpose
Obtainer is the agent-facing workflow for turning a data need into a production
training-data artifact. SearchAgent discovers hosted datasets, while the
registered Domain Data Acquisition WebAgent (domain_data_acquisition, legacy
alias webcrawler_dm) collects primary vertical-domain webpages as raw L1 data. DataMixer
is the only data-lake command surface for storage, ingest, processing,
indexing, sampling, recipe planning, export, snapshots, and lineage.
ObtainerCLI is the only supported end-to-end data workflow. Requests to clean, deduplicate, quality-filter, map, construct, or export a training dataset are Obtainer requests and must stay in the ObtainerCLI/DataMixer workflow through the final artifact.
When a long-running Codex SDK loop receives an Analyzer report, failure taxonomy, training recipe, or next-iteration data request, treat it as an Obtainer input, not a generic coding task:
- Identify whether the report needs dataset acquisition, production export, or both.
- For acquisition/download/ingest, start the managed
dataset-acquisition-agentworker instead of manually driving SearchAgent/WebAgent/download/ingest from the outer Codex context. - Poll worker status and decide whether to resume the same worker or start a fresh worker.
- Run the mandatory DataFlowAgent post-processing stage (
dm dataflow agent-run), which materializes the L4 dataset (quality, decontamination, deduplication, normalization, safety, and post-training validity). - Only after the DataFlowAgent run completes and the final L4 dataset scale
meets the recipe target, start the managed
sft-export-agentworker for production SFT outflow. If the user explicitly specifies an L3 export, the L4 gate is waived and L3 data may be exported directly once the lake volume/mix/quality gates pass. - Report warehouse path, datasets, record counts, recipe/export artifacts, lineage, manifests, and snapshots.
Main-Agent Use: Delegate to the Obtainer Orchestrator
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.
- 1mo ago First seen · 766 lines · 190 tokens per session scan A 5f3415a480e3
obtainer is a skill published in the GitHub repository OpenDCAI/Dataflow-LoopAI (29 stars, last pushed 16d ago), licensed Apache-2.0. It adds 190 tokens to every session and 9,889 once invoked, about $0.0008 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 2026-08-30.
Other skills, from other repositories
install-openviking-memory
Install and configure the OpenViking long-term memory plugin for OpenClaw via natural conversation. Once installed, the plugin automatically captures facts from chats and recalls relevant context before each reply (auto-capture + auto-recall, cross-session). Covers prerequisites, install through OpenClaw's plugin…
knowledge-distillation
Compile one or more OpenViking knowledge bases or document collections into topic-organized, evidence-grounded high-level knowledge, including cross-source findings, trends, changes, drivers, comparisons, implications, and uncertainties. Use with ov compile when the user asks to distill or synthesize a knowledge base…
knowledge-graph
Compile documents, notes, web content, transcripts, research materials, or code repositories into an evidence-grounded, visualization-ready knowledge graph with semantically typed entity nodes, statement-level provenance, and typed relationship edges. Use with ov compile to create or incrementally refresh entities/.md…
seo-llm
Use when optimizing content for LLM-powered search engines (ChatGPT, Perplexity, Gemini, Claude, Bing AI, Qwen), implementing RAG optimization, prompt engineering for search visibility, semantic SEO, and ensuring content ranks highly in AI-driven search results. Includes techniques for ChatGPT SEO, Perplexity…
llm-integrations
Use when integrating LLM providers (OpenAI, DeepSeek, OpenRouter, Anthropic, Google), configuring API keys, optimizing costs, implementing rate limiting, and managing LLM usage across projects. Includes best practices for cost optimization and API management. Based on OpenAI, Anthropic, Google, and other LLM provider…
qwen-agent
Use when developing Qwen-specific agents, implementing Qwen model integrations, optimizing Qwen performance, and building agent workflows with Qwen models. Includes Qwen API, Qwen-Plus, Qwen-Turbo, and best practices. Based on Qwen official documentation and agent development best practices.