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 build-with-dhiraj/ai-workflow-framework-portability-kit --skill fact-checkgit clone --depth 1 https://github.com/build-with-dhiraj/ai-workflow-framework-portability-kitWrote 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/build-with-dhiraj/ai-workflow-framework-portability-kit/fact-check)<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/fact-check"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/fact-check/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/build-with-dhiraj/ai-workflow-framework-portability-kit/fact-check"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/fact-check.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 | $0.00030 | $0.13052 |
| Opus 5.5 | $0.00012 | $0.05221 |
| Sonnet 5.5 | $0.00006 | $0.02610 |
| Haiku 4.5 | $0.00003 | $0.01305 |
Grade A, and why
fact-check 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 9d 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 — 1,110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fact-Check Skill v2.1
An advanced semi-automated fact-checking and disinformation detection system that produces visual Fact-Check Cards. Combines the SIFT methodology (Stop, Investigate, Find, Trace), the CRAAP test (Currency, Relevance, Authority, Accuracy, Purpose), prebunking/inoculation science, and claim decomposition with multi-language source triangulation, manipulation technique detection, origin tracing, counterfactual analysis, and a comprehensive educational component.
Grounded in research: fact-check labels reduce belief in false claims by ~18% (Clayton et al., 2020), accuracy prompts reduce sharing of false news by 15-20% (Pennycook & Rand, 2021), and prebunking videos improve manipulation recognition by ~5% even after a single viewing (van der Linden et al., 2022, Science Advances). This skill aims to maximize these effects through structured, transparent, and educational analysis.
Operating Modes
This skill operates in four modes depending on user intent:
Mode 1: Standard Fact-Check (default)
User provides content (text, URL, or image) and asks for verification. Runs the full 11-step pipeline. Produces a complete Fact-Check Card.
Mode 2: Comparison Mode
User provides two sources/texts/URLs on the same topic and asks for comparison. Runs decomposition on both, cross-references claims, identifies where they agree/disagree, evaluates which is more credible and why. Produces a Comparison Fact-Check Card.
Mode 3: Prebunking Briefing
User asks about current disinformation narratives on a topic (e.g., "What false narratives are circulating about vaccines/elections/energy?"). Searches for active disinformation campaigns on the topic, summarizes the most common false narratives, explains the manipulation techniques used, and provides preemptive defense tips. Produces a Prebunking Briefing Card.
Mode 4: Quick Check
User asks a simple yes/no verification question (e.g., "Is it true that X?"). Runs an abbreviated pipeline (Steps 1, 2, 6, 7 only). Skip Steps 3-5 and 8-9.
What ships with it
6 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.
- 9d ago First seen · 1,110 lines · 30 tokens per session scan A b19eb227ae69
fact-check is a skill published in the GitHub repository build-with-dhiraj/ai-workflow-framework-portability-kit (4 stars, last pushed 9d ago), licensed MIT. It adds 30 tokens to every session and 13,052 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-21.
Other skills, from other repositories
pith-tour
Interactive guided tour of Pith. Walks the user through each feature experientially — one step at a time, hands-on, using their actual project. Resumable. Skippable. State tracked via tour.py.
teach
Teach the user a new skill or concept, within this workspace.
canvas
Fetch Canvas LMS courses and assignments via API token.
huggingface-vision-trainer
Train object-detection, image-classification, or SAM segmentation models on Hugging Face Jobs. Use for vision fine-tuning and evaluation; use huggingface-llm-trainer for language models.
insecure-defaults
Detects fail-open insecure defaults (hardcoded secrets, weak auth, permissive security) that allow apps to run insecurely in production. Use when auditing security, reviewing config management, or analyzing environment variable handling.
ai-slop
Operational rubric that turns "don't make AI slop" into observable properties, severity levels, evidence requirements, and repair actions for interface design. Use as the reference rubric when building or reviewing marketing sites, product interfaces, dashboards, portfolios, or e-commerce pages, especially alongside…