Mintlify Docs
mintlify.com backs 9 GrowthDex tactics. This page exists so readers and crawlers can follow the evidence trail from source to tactic.
Original source
Cited GrowthDex tactics
- One-command skill install from docs URLTeach users to install the product's skill straight from the docs URL with one command so discovery turns into first use while intent is still warm.
- Skill frontmatter with compatibility and tool constraintsUse `skill.md` frontmatter to tell agents which environment assumptions matter and which tools are allowed before they start making things up.
- Agent-skills index for multi-workflow productsIf the product has several workflows, publish separate skill files plus a `/.well-known/agent-skills/index.json` manifest instead of forcing one giant capability blob.
- Agent-skills manifest with sha256 integrityPublish `/.well-known/agent-skills/index.json` with a digest for each skill so agents can verify they fetched the right instructions before they act on them.
- llms discovery headers on every pageAdd `Link` and `X-Llms-Txt` headers to normal page responses so an agent can find your machine-readable corpus before it starts crawling blindly.
- llms-full single-file context exportPublish an `llms-full.txt` file that bundles the important documentation corpus into one fetch for agents that work better with a single large context payload.
- Reverse-proxy forwarding for agent skill pathsIf docs sit behind a proxy or custom domain, forward `/skill.md`, `/.well-known/skills/*`, and `/.well-known/agent-skills/*` instead of letting the discovery layer die at the edge.
- Root skill.md for product capability discoveryPublish a root `/skill.md` that tells agents what your product can do, which inputs it needs, and which constraints matter instead of forcing them to infer capabilities from scattered docs.
- Well-known llms aliases for agent compatibilityServe `/.well-known/llms.txt` and `/.well-known/llms-full.txt` alongside the root files so agents that follow the well-known convention can discover your corpus without guessing.