Borrowing it
Nothing to install: this file belongs to JoshuaRamirez/advanced-prompting-engine. 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/JoshuaRamirez/advanced-prompting-engine/main/.claude/commands/research-semantic-improvement.mdgit clone --depth 1 https://github.com/JoshuaRamirez/advanced-prompting-engineWrote 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/joshuaramirez/advanced-prompting-engine/research-semantic-improvement)<a href="https://agentmods.dev/commands/joshuaramirez/advanced-prompting-engine/research-semantic-improvement"><img src="https://agentmods.dev/badge/commands/joshuaramirez/advanced-prompting-engine/research-semantic-improvement/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/joshuaramirez/advanced-prompting-engine/research-semantic-improvement"><img src="https://agentmods.dev/badge/commands/joshuaramirez/advanced-prompting-engine/research-semantic-improvement.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 | $0.00000 | $0.00928 |
| Opus 5 | $0.00000 | $0.00464 |
| Sonnet 5 | $0.00000 | $0.00186 |
| Haiku 4.5 | $0.00000 | $0.00093 |
Grade A, and why
research-semantic-improvement 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 12d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research: Semantic Bridge Improvement
Design a research plan to advance a philosophical intent measurement instrument from its current literary text benchmark score to 20/20 across Shakespeare, Genesis, Marx, MLK, Newton, Aristotle, Tao Te Ching, and Descartes. The instrument embodies a twelve-face geometric manifold where each face is a 12x12 grid with constitutive character on the x-axis and dispositional orientation on the y-axis. Its parser inverts three inference layers — axis meta-meaning, polarity convention, and sub-dimensions — by projecting intent tokens onto pre-computed GloVe 100d direction vectors. Remaining failures stem from: word polysemy where context determines domain, centroid broadness where some faces attract generic vocabulary, and vocabulary gaps where literary terms fall outside the word bridge. The research plan must systematically evaluate: (1) multi-sense static embeddings that provide sense-specific vectors without runtime neural inference — assess LMMS, sensEmbed, DeConf, and AutoExtend for GloVe compatibility and numpy-only resolution; (2) smooth inverse frequency sentence composition (Arora et al. 2017) as a replacement for simple word averaging — assess whether the principal component removal step sharpens face discrimination; (3) counter-fitting with antonym-only constraints and no face-cohort attraction to avoid centroid destabilization; (4) product quantization or dimensionality-indexed vocabulary expansion to cover 50K words within 10MB artifact budget; (5) character n-gram approximation for out-of-vocabulary literary terms using pre-computed prefix/suffix tables. For each technique, provide: expected improvement rationale, mathematical formulation, numpy implementation sketch, artifact size estimate, and risk assessment. Rank by expected impact on the specific benchmark failures.
Constraints
- Runtime: numpy only. No torch, tensorflow, onnxruntime, spacy, gensim, nltk.
- Artifacts: Ship with pip package. Under 10MB total.
- Build time: Any tool on developer machine. GloVe 6B already cached.
- Architecture: GeometricBridge loads pre-computed numpy arrays, intent parser tokenizes + looks up + averages. Improvements produce better arrays, not change runtime flow.
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.
- 12d ago First seen · 55 lines · 0 tokens per session scan A 96a0ee595f31
research-semantic-improvement is a command published in the GitHub repository JoshuaRamirez/advanced-prompting-engine (0 stars, last pushed 15d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 928 tokens. 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 commands, from other repositories
verify-math
Verify a self-authored mathematical result end to end by routing claims across adversarial review, numerical falsification, symbolic or CAS checks, and Lean, then aggregating one report. Use when a theorem, proposition, conjecture, or paper-wide mathematical argument needs the appropriate combination of verification…
replication-package
Scaffold or audit a social-science replication package at a target directory, and audit the manuscript and its archived research objects against FAIR principles.
diff
Quantitative volume comparison between a CadQuery model and a reference STEP file.
simulation-calibrator
Test and refine simulation accuracy with validation loops, bias detection, and continuous improvement frameworks.
arg-diagram
ARG academic-paper diagram mode — standalone structural & conceptual diagram generation.
graphite-morphology-classify
Classify graphite in a cast-iron micrograph per ASTM A247 / ISO 945-1, quantify nodularity, and read the matrix — the single most diagnostic observation in a cast-iron case.