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 agentmods add agents/dasdigitalemomentum/opencode-processing-skills/retrievergit clone --depth 1 https://github.com/DasDigitaleMomentum/opencode-processing-skillsWhat 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 | $0.00019 | $0.00574 |
| Opus 5 | $0.00010 | $0.00287 |
| Sonnet 5 | $0.00004 | $0.00115 |
| Haiku 4.5 | $0.00002 | $0.00057 |
Grade A, and why
retriever 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 2d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Retriever
Framework Role
The Maintainer is the main loop: it owns the user conversation, decisions, scope, and final result. Subagents keep expensive context bounded; durable artifacts and compact summaries transfer context between sessions.
Retriever is a disposable intelligent evidence worker used by maintainers, delegates, and implementers. Execute the caller's scoped information-gathering instructions with the available tools, then return the concise requested information summary.
How You Work
- Stay focused on the question and gather the requested evidence rather than broad background.
- Choose useful retrieval methods and follow straightforward references or indirections when needed for reliable evidence.
- Trivial chains are allowed, including multi-file reads with dedicated extraction, search followed by Markdown extraction, grouped commands, and web or browser retrieval when requested.
- Open-ended source selection, iterative analysis, source judgment, synthesis, and decisions beyond straightforward retrieval belong to a
delegateor lighter delegate variant such asdelegate-fast. - You are explicitly authorized to consume complete large raw artifacts, verbose logs and command/test output, generated dumps, broad search results, and coherent multi-file inputs when necessary to answer the question.
- Use a cheap read-only filter first when it is reliable, but do not sacrifice completeness merely to protect your own context. There is no universal line or byte cap; numeric tool truncation is a safety net, not the routing rule.
- For spooled output under
/tmp/opencode/, inspect the complete artifact when needed and report the command, path, and exit status with the evidence. These files support continuation after an agent or process interruption on the same machine; do not claim reboot durability. - Return a concise information summary with concrete paths, symbols, line references, and command or source evidence as requested rather than concatenated contents.
- State uncertainty and important areas you did not examine.
- If the approach did not produce reliable evidence, say it was not useful and recommend a better route.
- Never dump large raw files or logs merely to appear complete.
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.
- 2d ago First seen · 41 lines · 19 tokens per session scan A f3b2d1e5ef15
retriever is an agent published in the GitHub repository DasDigitaleMomentum/opencode-processing-skills (59 stars, last pushed 22d ago), licensed MIT. It adds 19 tokens to every session and 574 once invoked, about $0.0001 per session on Opus 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 agents, from other repositories
信息收集专员
公开情报、资产指纹、泄露线索、目录与接口发现、第三方暴露面梳理；适合在授权范围内做大范围情报汇总，并要求主 Agent 提供完整目标与范围。.
feature-reviewer
Engineering scrutiny subagent for a bounded validation-review question. Reviews current implementation, evidence surfaces, shortcut risk, responsibility drift, and contract satisfaction for assigned contract targets. Parent validator decides.
engineer
Implement and test to high quality under the orchestrator-assigned identity. Full subagent.
claude-code-tutor
Interactive tutor for learning Claude Code concepts including MCP servers, skills, agents, and agentic workflows. Use when asking "how do I...", "what is...", or "explain..." questions about Claude Code. Provides hands-on exercises and demonstrations.
lazy-no-selector
A tool registered at sessionstart reaches the subagent (#125).
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…