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.
git clone --depth 1 https://github.com/amirkiarafiei/open-scholar-peerWrote 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/commands/amirkiarafiei/open-scholar-peer/5-osp-qa)<a href="https://agentmods.dev/commands/amirkiarafiei/open-scholar-peer/5-osp-qa"><img src="https://agentmods.dev/badge/commands/amirkiarafiei/open-scholar-peer/5-osp-qa/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/commands/amirkiarafiei/open-scholar-peer/5-osp-qa"><img src="https://agentmods.dev/badge/commands/amirkiarafiei/open-scholar-peer/5-osp-qa.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 | $0.00020 | $0.01042 |
| Opus 5 | $0.00010 | $0.00521 |
| Sonnet 5 | $0.00004 | $0.00208 |
| Haiku 4.5 | $0.00002 | $0.00104 |
Grade A, and why
5-osp-qa 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 11d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tool capability: This tool supports subagents. The Query Agent MUST delegate each question to
osp-answer-generator-agentas a subagent with a fresh, minimal context bundle. Do NOT use self-reflection.
/5-osp-qa — Multi-Aspect Q&A Engine
For each criterion in session.json.qa_criteria[], generate N probing Q&A pairs (N = qa_pairs_per_criterion, default 2). The Query Agent runs in the main thread; the Answer Generator runs as a subagent (or self-reflects on Antigravity).
Activation
Invoke the osp-query-agent skill (main thread). The Query Agent will spawn osp-answer-generator-agent per question.
Prerequisites
phases.summary.status,phases.literature.status,phases.historian.status,phases.baseline_scout.statusall"completed".qa_criteria[]is non-empty insession.json.
Step 0 — Resource check and pair count (run BEFORE any Q&A work)
-
Read
session.json. Countqa_criteria[]items (call it C). -
Read
qa_pairs_per_criterionfromsession.json(default 2 if absent). -
Print the resource estimate:
⚠️ Q&A Engine — resource estimate Criteria: C Pairs/criterion: N (currently set in session.json) Total subagent calls: C × N = <total> Estimated time: ~<total × 45s> at typical API latency Pair count guide: 2 — quick scan, catches the most obvious issues (default) 5 — thorough coverage, good for most reviews 10 — exhaustive, suitable for high-stakes decisions -
Ask the user: "How many Q&A pairs per criterion? Press Enter for [N] or type a number (2–10):"
-
If the user enters a number, update
session.json.qa_pairs_per_criterionto that value and use it. If the user presses Enter, use the existing value.
Mode selection
- Subagent mode (default): Claude Code, Cursor, Gemini CLI, GitHub Copilot CLI. The Query Agent delegates each question to the Answer Generator as a subagent with a fresh, minimal context bundle.
- Self-reflection mode (Antigravity only): The Query Agent uses strict turn markers (
=== Query Agent === ... === END === === Answer Generator === ...) within the main context window.
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.
- 11d ago First seen · 86 lines · 20 tokens per session scan A 62dd3c053114
5-osp-qa is a command published in the GitHub repository amirkiarafiei/open-scholar-peer (27 stars, last pushed 1mo ago), licensed MIT. It adds 20 tokens to every session and 1,042 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 commands, from other repositories
ars-mark-read
ARS /ars-mark-read — record a user-attested reading signal for one or more citation keys.
ars-cache-invalidate
ARS /ars-cache-invalidate — drop cached verification entries for a citation key.
ars-unmark-read
ARS /ars-unmark-read — rescind a user-attested reading signal for one or more citation keys.
ars-revision-coach
ARS academic-paper revision-coach — peer-review roadmap or source-accounted real-committee response skeleton.
ars-disclosure
ARS academic-paper disclosure mode — venue applicability/status bundle or policy-anchor render.
ars-rebuttal-audit
ARS academic-paper rebuttal-audit mode — QA an existing rebuttal draft against reviewer comments.