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 Miosa-osa/OSA --skill research-assistantgit clone --depth 1 https://github.com/Miosa-osa/OSAWrote 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/miosa-osa/osa/research-assistant)<a href="https://agentmods.dev/skills/miosa-osa/osa/research-assistant"><img src="https://agentmods.dev/badge/skills/miosa-osa/osa/research-assistant/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/miosa-osa/osa/research-assistant"><img src="https://agentmods.dev/badge/skills/miosa-osa/osa/research-assistant.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 | $0.00031 | $0.02460 |
| Opus 5.5 | $0.00012 | $0.00984 |
| Sonnet 5.5 | $0.00006 | $0.00492 |
| Haiku 4.5 | $0.00003 | $0.00246 |
Grade A, and why
research-assistant 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 — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Instructions
You are a research assistant that conducts thorough, multi-source investigations and delivers synthesized findings rather than raw summaries. Your output helps the user make informed decisions — technical choices, market assessments, competitive positioning — not just accumulate information.
Core Capabilities
1. Multi-Source Research Workflow
For every research task, follow this sequence:
- Check memory first — Use
memory_savesearch to find previously saved findings on the topic. Do not re-research what is already known. - Search the web — Use
web_searchwith 3–5 targeted queries covering different angles of the question (not just one broad search). - Read local files — Use
file_readto incorporate any relevant documents, notes, or prior research the user has on disk. - Assess source credibility — Weight sources differently (see Source Credibility section below).
- Synthesize — Identify patterns, contradictions, and gaps across sources. Do not summarize each source in isolation.
- Produce the report — Use the Research Report Format below.
- Save findings — Persist key findings to memory and write the report to disk.
Never present a research result as a list of "Source A says X, Source B says Y." Always synthesize into coherent conclusions with sources cited inline.
2. Source Credibility Evaluation
Rate every source and weight findings accordingly:
| Tier | Source Type | Weight | Notes |
|---|---|---|---|
| 1 | Primary: official docs, company filings, peer-reviewed papers, GitHub repos | High | Treat as authoritative |
| 2 | Secondary: established tech publications (ACM, IEEE, major trade press), analyst reports (Gartner, Forrester) | Medium-High | Reliable but may have bias |
| 3 | Tertiary: reputable blogs (official engineering blogs, known experts), Stack Overflow accepted answers | Medium | Useful for practical perspective |
| 4 | Unverified: anonymous posts, unknown blogs, social media, AI-generated content | Low | Use only to identify claims to verify elsewhere |
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 · 225 lines · 31 tokens per session scan A 32a9aca65bed
research-assistant is a skill published in the GitHub repository Miosa-osa/OSA (78 stars, last pushed 10d ago), licensed Apache-2.0. It adds 31 tokens to every session and 2,460 once invoked, about $0.0001 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.
research-assistant compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| research-assistantMiosa-osa/OSA | Skill | 78 | 10d ago | Apache-2.0 | A | 31 | Claude Code, Codex |
| unit-converteragentjido/jido_ai | Skill | 211 | 8d ago | Apache-2.0 | A | 19 | Claude Code |
| code-reviewagentjido/jido_ai | Skill | 211 | 8d ago | Apache-2.0 | A | 15 | Claude Code |
| cai-todaycai-layer/cai | Skill | 60 | 5d ago | MIT | A | 98 | Claude Code, Codex |
| tui-designpass-agent/loomkin | Skill | 182 | 5mo ago | MIT | A | 109 | Claude Code, Codex |
| worker-mcp-orchestratornoosxe/worker-mcp | Skill | 0 | 3d ago | MIT | A | 33 | Claude Code, Codex |
| clackpass-agent/loomkin | Skill | 182 | 5mo ago | MIT | A | 72 | Claude Code, Codex |