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 AutoRFP/mcp --skill content-library-contradictionsgit clone --depth 1 https://github.com/AutoRFP/mcpWrote 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/autorfp/mcp/content-library-contradictions)<a href="https://agentmods.dev/skills/autorfp/mcp/content-library-contradictions"><img src="https://agentmods.dev/badge/skills/autorfp/mcp/content-library-contradictions/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/autorfp/mcp/content-library-contradictions"><img src="https://agentmods.dev/badge/skills/autorfp/mcp/content-library-contradictions.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 | $0.00184 | $0.02064 |
| Opus 5 | $0.00092 | $0.01032 |
| Sonnet 5 | $0.00037 | $0.00413 |
| Haiku 4.5 | $0.00018 | $0.00206 |
Grade A, and why
content-library-contradictions 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 10d 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Content Library Contradictions
Why this skill exists
An AutoRFP.ai content library is the single source of truth that feeds every RFP, DDQ, and security questionnaire response. When two answers in the library disagree with each other, or when an old answer contradicts the company's current state, the response engine can pull the wrong one into a live deal and put incorrect information in front of a buyer. This is one of the highest-impact hygiene problems a content team can fix, and it compounds: every contradiction left in place will be pulled into future deals.
This skill audits the user's library for three classes of contradiction:
- Conflicting facts — two answers state different numbers, names, or claims for the same thing (e.g. one answer lists one supported language count, another lists a different one).
- Outdated facts — an old answer contradicts the company's current state (e.g. an answer says the company doesn't have a capability the company has since launched).
- Internal capability contradictions — two answers describe mutually exclusive capabilities (e.g. one says data is hosted in a single region, another references multi-region deployments).
The skill flags contradictions; it does not auto-edit content. The output is an Excel file the user reviews to decide which version of each fact is correct.
When to trigger
Trigger whenever the user asks about contradictions, inconsistencies, conflicting answers, or library health. Don't confuse with rfp-contradiction-checker (which scans a finished RFP response, not the library).
Inputs you need from the user
Before pulling any data, ask the user clarifying questions using the AskUserQuestion tool. Even if the user's request seems clear, the audit is much more useful when scoped. Ask:
- Scope — audit the whole library, or focus on a topic area? Common topic scopes include security, integrations, AI/architecture, company information, pricing, support, implementation. A scoped audit is faster and produces a more actionable list.
- Tag prioritisation — call
list_tagsfirst to discover the org's actual taxonomy. Then ask: which tag categories matter most for this audit? Use the discovered tags as option labels — don't make up tags that don't exist in the user's library. - Ground-truth anchors — are there facts the user knows the current truth for? Typical anchor categories for a B2B SaaS company are: where customer data is hosted (regional architecture), current team size, AI/LLM model stack, certifications held, integrations supported, current customer roster used as proof, uptime / SLA commitments, pricing model. These anchors become the reference points for detecting outdated facts. If the user can't easily list anchors, derive a working hypothesis from the most recently updated content and flag this assumption clearly in the output.
- Time window — include all content, or only content older than N months? Recent content is less likely to be outdated but can still conflict with peer answers.
- Severity threshold — flag everything, or only contradictions in content that's actively being pulled into deals (usageCount above a threshold)? High-usage contradictions are the actionable subset.
What ships with it
3 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.
- 10d ago First seen · 114 lines · 184 tokens per session scan A 41b5bd5d38ae
content-library-contradictions is a skill published in the GitHub repository AutoRFP/mcp (0 stars, last pushed 3mo ago), licensed MIT. It adds 184 tokens to every session and 2,064 once invoked, about $0.0009 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…