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/midnightdarling/collate/real-thesisnpx skills add MidnightDarling/collate --skill real-thesisgit clone --depth 1 https://github.com/MidnightDarling/collateWrote 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/midnightdarling/collate/real-thesis)<a href="https://agentmods.dev/skills/midnightdarling/collate/real-thesis"><img src="https://agentmods.dev/badge/skills/midnightdarling/collate/real-thesis.svg" alt="Measured on agentmods" 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 | $0.00039 | $0.00444 |
| Opus 5 | $0.00019 | $0.00222 |
| Sonnet 5 | $0.00008 | $0.00089 |
| Haiku 4.5 | $0.00004 | $0.00044 |
Grade A, and why
real-thesis 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 6d 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.
What it actually says
real-thesis
The work this skill is for
Many papers are written around a safer surface topic. The explicit claim is the outer layer; the real pressure lives deeper in the structure — what returns, what gets omitted, where the densest citations accumulate, where the conclusion lands with disproportionate force.
This skill excavates that buried centre.
What to look for
When the surface argument unfolds, pay attention to:
- where the author returns repeatedly
- what is
略而不论or passed over in silence - where the conclusion lands and what it cannot let go
- where citations grow abnormally dense
- where footnotes work harder than the body
The aim is not psychoanalysis for its own sake. The aim is to identify the thesis that, if spoken plainly, might reorganize the abstract, the structure, or even the title.
Input
Target: $ARGUMENTS
- workspace directory → read
<ws>/final.md .mdfile → read it directly- empty → ask what to read
Output
- workspace mode →
<workspace>/analysis/{stem}_real-thesis.md - single-file mode →
analysis/{stem}_real-thesis.md
Report layers:
- Surface topic — what the paper openly claims
- Floating concerns — what it repeatedly approaches but does not settle
- Candidate real theses — one to three, each anchored in evidence: returns, elisions, citations, footnotes
- A question the author dare not ask themselves — left open, not answered for them
Guardrails
- Never overwrite the source.
- Do not upgrade a mood into a thesis without evidence.
- If the paper is actually direct and nothing deeper is hiding, say so.
- Precision matters more than theatrical boldness.
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.
- 6d ago First seen · 59 lines · 39 tokens per session scan A 7e0256e7db4c
real-thesis is a skill published in the GitHub repository MidnightDarling/collate (6 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 39 tokens to every session and 444 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
parallel-feature-development
Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns for multi-agent implementation. Use this skill when decomposing a large feature into independent work streams, when two or more agents need to implement different layers of the same system…
nft-standards
Implement NFT standards (ERC-721, ERC-1155) with proper metadata handling, minting strategies, and marketplace integration. Use when creating NFT contracts, building NFT marketplaces, or implementing digital asset systems.
cost-optimization
Optimize cloud costs across AWS, Azure, GCP, and OCI through resource rightsizing, tagging strategies, reserved instances, and spending analysis. Use when reducing cloud expenses, analyzing infrastructure costs, or implementing cost governance policies.
istio-traffic-management
Configure Istio traffic management including routing, load balancing, circuit breakers, and canary deployments. Use when implementing service mesh traffic policies, progressive delivery, or resilience patterns.
postgresql-table-design
Use this skill when designing or reviewing a PostgreSQL-specific schema. Covers best-practices, data types, indexing, constraints, performance patterns, and advanced features.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.