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 skills/yogsoth-ai/de-anthropocentric-research-engine/optimization-loopnpx skills add yogsoth-ai/de-anthropocentric-research-engine --skill optimization-loopgit clone --depth 1 https://github.com/yogsoth-ai/de-anthropocentric-research-engineWhat 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 | $0.00065 | $0.02343 |
| Opus 5 | $0.00032 | $0.01171 |
| Sonnet 5 | $0.00013 | $0.00469 |
| Haiku 4.5 | $0.00006 | $0.00234 |
Grade A, and why
optimization-loop 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 yesterday.
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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
optimization-loop
You are the optimizer. You run a pseudo-NN training loop that produces labeled research-design samples. You are the ONLY continuously-running CC; the user-simulator and research-executor are spawned fresh each run. Disk is the only source of truth — your memory is not.
§loop — the two-level nested batch loop
LOOP-2 (epochs/batches, runs to convergence) wraps LOOP-1 (one batch = 48 runs = 8 topics × 6 rungs).
Epoch start (once):
python ../../scripts/new_run_id.py --runs-root runs→ capturerun_id. Emit tracerun_start.
Batch start (each batch):
2. batch_id = the highest batch-N in runs/<run_id>/weights/ (cold start =
batch-0; do NOT pre-increment). Emit batch_start.
3. Generate configs from the current weights: call
../../generator/gen_configs.py main(out_dir, w) (it is main(out_dir, w),
NOT an argparse CLI — invoke via a one-line python -c that imports it and
passes the loaded weights/<batch_id>.json). M1 note: gen_configs names
configs config_{rung}.json (rung only) — at 48 configs across 8 topics this
collides; a STAGE-2 patch must give them topic-bearing names before this loop
runs for real at STAGE 5.
Per run (8 topics × 6 rungs = 48):
4. Spawn the user-simulator CC (see §tools). Inject its research_config.
5. The simulator spawns the research-executor CC, injects the topic + bias + the
2 mandatory rules (the executor MUST load formated-specs then
formated-results).
6. Simulator ↔ executor talk for the card's turn budget. Emit dialogue_turn
per exchange.
7. python ../../scripts/save_transcript.py --logs-dir <REQUIRED> --cwd <exec-cwd> --sample <sample> --out runs/<run_id>/transcripts/<sample>.md. --logs-dir
is REQUIRED, no default (privacy red line).
8. python ../../scripts/concat_triple.py --run-dir runs/<run_id> --sample <sample>.
9. python scripts/run_codex_loss.py loss1 --transcript … --config … --skill-md ../injection-fidelity/SKILL.md --out runs/<run_id>/loss/<sample>.loss1.json.
Emit rung_done.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 167 lines · 65 tokens per session scan A a5f39bd077dc
optimization-loop is a skill published in the GitHub repository yogsoth-ai/de-anthropocentric-research-engine (405 stars, last pushed 6d ago), licensed Apache-2.0. It adds 65 tokens to every session and 2,343 once invoked, about $0.0003 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 skills, from other repositories
Deep Research
Produce a deep, structured research report on a topic: decompose into key dimensions, analyze each with evidence and reasoning, synthesize cross-cutting insights, and surface open questions. Use for deep research, analysis, and literature/landscape reviews.
omnisci
Run OmniScientist end to end in the OmniScientist CLI using DeepSeek V4 Flash. Turn raw research data (images, signals, audio, video, 3-D, tables, or graphs) and an open direction into perceived evidence, a falsifiable hypothesis, recorded analysis, real citations, a gated candidate paper, PDF, and Overleaf bundle.…
literature-reviewer-skill
根据用户提供的论文主题，进行系统性中英文文献回顾（Literature Survey）。 采用8阶段工作流，支持CNKI、Web of Science、ScienceDirect等主流数据库， 无需API配置，通过浏览器自动化获取文献信息。 输出包含GB/T 7714-2015引文、标题、摘要的Markdown文档。 当用户提到"文献回顾"、"文献综述"、"帮我找文献"、"中英文文献搜索"、"写综述"等关键词时触发。 NOTE: The 'examples/' folder contains sample outputs and is not included in skill installation.
academic-research-suite
ARS-Codex workflows for research, academic writing, manuscript review, research-to-paper pipelines, and experiment planning. Use when the user asks for deep research, literature review, systematic review, meta-analysis, research question refinement, academic paper drafting, paper revision, citation or integrity…
paper-fetch
Add academic papers to Zotero via MCP (DOI/PMID/title); fetch full-text PDFs through a publisher-aware route ladder (Unpaywall → Elsevier/Wiley/Springer TDM → institutional proxy → SFX); PDFs land in a linked-files folder (ZotMoov), never uploaded to Zotero cloud. Also verifies that a downloaded PDF really is that…
lit-review-assistant
Search, summarize, and synthesize economics literature.