Borrowing it
Nothing to install: this file belongs to morluto/jacobian. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/morluto/jacobian/main/.agents/skills/recent-conjecture-evaluations/SKILL.mdgit clone --depth 1 https://github.com/morluto/jacobianWrote 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/morluto/jacobian/recent-conjecture-evaluations)<a href="https://agentmods.dev/skills/morluto/jacobian/recent-conjecture-evaluations"><img src="https://agentmods.dev/badge/skills/morluto/jacobian/recent-conjecture-evaluations/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/morluto/jacobian/recent-conjecture-evaluations"><img src="https://agentmods.dev/badge/skills/morluto/jacobian/recent-conjecture-evaluations.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 | $0.00022 | $0.00625 |
| Opus 5 | $0.00011 | $0.00313 |
| Sonnet 5 | $0.00004 | $0.00125 |
| Haiku 4.5 | $0.00002 | $0.00063 |
Grade A, and why
recent-conjecture-evaluations 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 4d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recent Conjecture Evaluations
Use recently resolved conjectures to probe Jacobian reliability. The outcome is an evidence-backed diagnosis, not a collection of solved examples. This skill owns source selection, deterministic replay, optional model comparisons, and attribution; Harbor packaging is a separate workflow.
Choose the requested mode
- Probe: read probe workflow for one source cycle, including source gating, independent oracle, current-main replay, and reporting.
- Review: inspect the completed cycle against source gating, the oracle and frozen inputs, and the applicable scoring rules. Use report fields to identify missing evidence. Reproduce disputed deterministic claims where useful; do not repeat model arms.
- Coordinate: search saved reports, reservations, and issue/PR ownership for the source and root mechanism. Use the inventory helper shown below and the ownership rules in action policy. Do not begin a new probe merely to answer a coordination question.
python .agents/skills/recent-conjecture-evaluations/scripts/search_inventory.py \
"source or root-cause phrase" outputs benchmarks/results
Preserve the evidence boundary
Bind prompts, payloads, and oracles to the exact intended input. Reconstruct gold independently of evaluated model arms and distinguish source-supplied replay from an independent oracle. Record precise source status and dates. Computation, verification, imported theorems, and unproved claims establish different things. Timeouts, unavailable operations, and missing witnesses are non-conclusions. Never attribute a model's fallback or transcription error to Jacobian.
Audit new probes deterministically on current main before paid model calls. Run comparisons only when they resolve uncertainty that direct evidence cannot, with user-authorized cost boundaries and frozen control/treatment conditions. Do not weaken verification or create a benchmark-specific operation to make a probe pass. Before external action, apply the action policy within existing user authorization. This evaluation workflow produces at most a localized draft PR; merging is a separate user-authorized task.
What ships with it
7 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.
- 4d ago Changed · -94 lines · -64 tokens per session bf9fe9876ebc
- 9d ago First seen · 157 lines · 86 tokens per session scan A 1d5a62444ec8
recent-conjecture-evaluations is a skill published in the GitHub repository morluto/jacobian (188 stars, last pushed today), licensed MIT. It adds 22 tokens to every session and 625 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 skills, from other repositories
discriminating-experiment-design
Recognize when a benchmark cannot discriminate between methods, construct experiment families that could, sweep structural knobs without sweeping the decision itself, and report a tie that survives as the headline result. Use when all methods score the same on your benchmark, when designing a new evaluation world or…
experiment-audit
A review step that checks whether an experiment's reported results are supported by real data and a sufficiently broad test.
paper-search
Search, download, and read academic papers from 20+ sources (arXiv, PubMed, Semantic Scholar, CrossRef, etc). Use when the user asks to find papers, search for research, look up academic literature, download a paper PDF, or extract text from a paper.
flutter-mcp-toolkit-control
Drive a running Flutter app — tap, scroll, type, fill forms, hot-reload, navigate. Use when you need to interact with the UI.
harness-engineering-lifecycle
Design, implement, and integrate generalized validation harnesses across a producer-consumer boundary after a local harness contract exists. Use when refactoring custom validation CLIs/MCPs for large polyrepos, extending Steward across sibling repos, or deploying a local tool to a consumer project for dogfooding and…
flutter-mcp-toolkit-dogfood-iterations
Runs and records fluttertestapp dogfood iterations (toolqualityrubric, rundogfoodeval.sh, dogfoodwebeval.yaml). Use when scoring MCP/intentcall quality, appending iteration N, comparing regressions, or CI static/weekly eval gates.