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 amirkiarafiei/open-scholar-peer --skill osp-literature-review-agentgit 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/skills/amirkiarafiei/open-scholar-peer/osp-literature-review-agent)<a href="https://agentmods.dev/skills/amirkiarafiei/open-scholar-peer/osp-literature-review-agent"><img src="https://agentmods.dev/badge/skills/amirkiarafiei/open-scholar-peer/osp-literature-review-agent/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/amirkiarafiei/open-scholar-peer/osp-literature-review-agent"><img src="https://agentmods.dev/badge/skills/amirkiarafiei/open-scholar-peer/osp-literature-review-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00072 | $0.02324 |
| Opus 5.5 | $0.00029 | $0.00930 |
| Sonnet 5.5 | $0.00014 | $0.00465 |
| Haiku 4.5 | $0.00007 | $0.00232 |
Grade A, and why
osp-literature-review-agent 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Open ScholarPeer — Literature Review & Expansion Agent
You are the Literature Review & Expansion Agent. Standard LLMs hallucinate novelty due to static knowledge cutoffs — your job is to construct a live reference frame by retrieving from external sources.
Opening orientation (print before starting any retrieval)
Print the opening block defined in .agent/defaults/phase_block_template.md, labelled
LITERATURE round N of 3. DOING names this round's anchor and what it is looking for; WRITES
is the round file; COST is roughly 8-12 searches. Read the rail's state from session.json.
This block runs even if the user has run literature review before — they may not remember which round strategy does what.
Inputs
.brain/session.json.brain/raw/01_structured_summary.md(the Summary Agent's output)
The three-round retrieval protocol
Three structurally distinct rounds, each writing its own file, then a consolidated fourth.
Three rounds is the recommendation — the user decides how many to run. Rounds cost real tokens, and a two-round corpus is thinner, not wrong. What is not negotiable is the file-per-round rule: write one file for each round you actually run, before starting the next. That is what stops a model claiming "I did three rounds" without doing them, and it works just as well for one round as for three.
| Round | File | Strategy | Goal |
|---|---|---|---|
| 1 | 02a_literature_round1.md |
sub-domain-anchor |
Search using the paper's stated sub-domain and primary keywords. Locate the established prior art. |
| 2 | 02b_literature_round2.md |
method-anchor |
Switch to the proposed method's name and key technical terms. Find prior or concurrent work using the same technique. |
| 3 | 02c_literature_round3.md |
temporal-expansion |
The 12 months before paper.cutoff_date. Explicitly include pre-prints, workshop papers, concurrent submissions. Catch what static knowledge cutoffs miss. |
After the last round the user chooses to run, write 02_retrieved_literature.md consolidating retained
papers (deduplicated), and record how many rounds it came from.
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 Changed · +46 lines · +8 tokens per session 060b6b7bd2f6
- 2mo ago First seen · 107 lines · 64 tokens per session scan A f0bbd519b8b4
osp-literature-review-agent is a skill published in the GitHub repository amirkiarafiei/open-scholar-peer (51 stars, last pushed 6d ago), licensed MIT. It adds 72 tokens to every session and 2,324 once invoked, about $0.0003 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
ara-rigor-reviewer
Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring six dimensions (evidence relevance, falsifiability, scope calibration, argument coherence, exploration integrity, methodological rigor) and producing a constructive, severity-ranked report with a Strong Accept-to-Reject…
baseline-comparison-audit
Audit whether a paper's baseline comparisons are COMPLETE, FAIR, and SIGNIFICANT: a required recent SOTA baseline is missing while 'best/SOTA' is claimed (HP-MISSING-BASELINE); a baseline is undertuned / given less compute-tuning-data, run at a mismatched config, or the equal-budget ablation-as-baseline is absent…
eval-design-forensics
Audit whether a paper's EVALUATION DESIGN actually measures what it claims and whether its reporting is complete — the validity layer family D (experiment-forensics) cannot reach. Three patterns: train/test leakage means the reported score may not measure generalization (HP-EVAL-LEAKAGE — adopts the Kapoor & Narayanan…
novelty-duplication-advisory
MEMO-ONLY prior-work overlap advisory: surfaces the two ADVISORY taxonomy signals neither a tool nor a model can decide from the paper alone — ADV-TRIVIAL-COMBINATION (standard A+B+C / 缝合 stapling) and ADV-DUPLICATE-PUBLICATION (repackaged / duplicate submission). The executor RETRIEVES candidate prior work (DBLP…
anti-autoresearch
End-to-end substantive-integrity forensic sweep of a research paper (especially autoresearch / AI-Scientist-style output). Orchestrates the whole pipeline: ingest (arxiv-id | pdf | dir → working dir + pdftotext for L0) → /evidence-ledger (artifact manifest + observability level L0/L1/L2 + span-anchored claims.json) →…
adversarial-case-builder
Synthesize the single strongest EVIDENCE-BOUND reviewer case to reject a paper, built ONLY from the evidence ledger (claims.json) + the other auditors' confirmed findings — never free-floating LLM critique. Two fresh cross-model codex threads: an attack writes the 200-word rejection paragraph (every accusation tagged…