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 docxology/template --skill verificationgit clone --depth 1 https://github.com/docxology/templateWrote 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/docxology/template/verification)<a href="https://agentmods.dev/skills/docxology/template/verification"><img src="https://agentmods.dev/badge/skills/docxology/template/verification/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/docxology/template/verification"><img src="https://agentmods.dev/badge/skills/docxology/template/verification.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 | $0.00124 | $0.00712 |
| Opus 5 | $0.00062 | $0.00356 |
| Sonnet 5 | $0.00025 | $0.00142 |
| Haiku 4.5 | $0.00012 | $0.00071 |
Grade A, and why
reference-verification 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 5d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Descriptor — infrastructure/reference/verification
Module Overview
Given the references a manuscript claims to cite, prove (or disprove) that
each one exists and matches its cited metadata. This is the verification side of
the literature workflow; infrastructure/search/literature is the discovery
side.
Capabilities
- Resolve: DOI → Crossref then OpenAlex (cross-index triangulation); arXiv id → arXiv API; bare title → Crossref title search with a similarity floor.
- Classify (deterministic):
ok,mismatch,fabricated,unverifiable,unchecked,anachronism. - Offline-first: with
allow_network=False(default), consult only the SQLite cache and reportuncheckedon a miss — a skipped check never launders into a clean pass. - Persistent cache: SQLite, 90-day TTL, stores negative results so "fabricated" is fast and stable.
- Temporal integrity: flag citations dated after the manuscript's as-of year.
CLI
uv run python -m infrastructure.reference.verification verify references.bib
uv run python -m infrastructure.reference.verification verify references.bib --live --as-of-year 2026 --fail-on-issues
uv run python -m infrastructure.reference.verification verify references.bib --json
uv run python -m infrastructure.reference.verification cache-clear
Python API
from infrastructure.reference.verification import ReferenceResolver, ResolutionCache, verify_bibfile
resolver = ReferenceResolver(cache=ResolutionCache("cache.db"), allow_network=True)
report = verify_bibfile("references.bib", resolver, as_of_year=2026)
print(report.summary_line())
assert not report.has_blocking
See Also
resolver.py— DOI/arXiv/title resolutionverifier.py— classification + temporal integritycache.py— persistent SQLite resolution cacheinfrastructure/search/literature/SKILL.md— discovery sidedocs/prompts/manuscript-claim-verification/SKILL.md— workflow that drives this gate
What ships with it
9 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.
- 5d ago First seen · 62 lines · 124 tokens per session scan A 80eec63f997a
reference-verification is a skill published in the GitHub repository docxology/template (19 stars, last pushed today), licensed Apache-2.0. It adds 124 tokens to every session and 712 once invoked, about $0.0006 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 2026-09-03.
Other skills, from other repositories
aerospace-engineering-technician
Use when a task needs the judgment of an Aerospace Engineering and Operations Technologist/Technician — verifying an installed fastener's preload against a drawing's torque callout via the T=K·D·F relationship, reducing strain-gauge data from a structural proof-load test into stress and checking it against an…
agricultural-engineer
Use when a task needs the judgment of an agricultural engineer — sizing a center-pivot or drip irrigation system's peak capacity against crop water demand, computing lateral grain-bin wall pressure with Janssen's equation, sizing a waterway or tile-drainage system with Manning's equation and an NRCS design-storm…
anthropologist-archeologist
Use when a task needs the judgment of an Anthropologist/Archeologist working in cultural resource management (CRM) — scoping a Section 106 Phase I identification survey, designing a shovel-test-pit sampling strategy, evaluating National Register of Historic Places eligibility under Criteria A-D, interpreting…
automotive-engineering-technician
Use when a task needs the judgment of an Automotive Engineering Technician — setting up and instrumenting test equipment (strain-gauge bridges, thermocouples, load/torque sensors) to an engineer's written test plan, selecting DAQ sample rate and an SAE J211 CFC filter class for a vehicle test channel, verifying…
calibration-technician
Use when a task needs the judgment of a Calibration Technologist/Technician — computing a Test Uncertainty Ratio (TUR) and deciding whether it meets the 4:1 target, building a measurement uncertainty budget (Type A/Type B, GUM-style) for a calibration, tracing a reference standard's chain of custody to NIST/SI…
acoustical-engineer
Use when a task needs the judgment of an Acoustical Engineer — calculating room reverberation time (RT60) and specifying absorption/reflection treatment for a space, evaluating a wall or floor assembly's STC/IIC rating against a code or design target, diagnosing why a completed space fails its acoustic design intent…