Prepare a source document for wiki processing. Supports Markdown, PDF, DOCX, and PPTX sources. Returns the source content, TOC (for binary formats), related wiki pages, and suggestions. Features: - Incremental: Tracks previously ingested sources. If unchanged, returns cached status with derived w...
AI agents invoke wiki_ingest to trigger actions in MinerU Document Explorer. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
wiki_ingest triggers real processes with real consequences. An agent gone sideways doesn't fire it once — it starts dozens of builds, sends mass notifications, or burns through compute before anyone looks up.
Documented attack patterns abuse exactly the kind of access wiki_ingest gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MinerU Document Explorer, and nothing reaches the server without passing your rules. This is the rule we recommend for wiki_ingest:
{
"version": "1",
"default": "deny",
"tools": {
"wiki_ingest": {
"limits": [
{
"counter": "wiki_ingest_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}wiki_ingest stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Prepare a source document for wiki processing. Supports Markdown, PDF, DOCX, and PPTX sources. Returns the source content, TOC (for binary formats), related wiki pages, and suggestions. Features: - Incremental: Tracks previously ingested sources. If unchanged, returns cached status with derived wiki pages. Use force=true to re-ingest. - Multi-format: PDF/DOCX/PPTX sources include a structured TOC and format-specific metadata (page/slide/section counts). - Large docs: Bodies >50k chars are truncated with a suggestion to use doc_read for specific sections. This tool does NOT generate wiki pages — it gives you context to do so. After calling this, use doc_write with the source parameter to create wiki pages and record provenance. Workflow: 1. Call wiki_ingest with a source document 2. Read the returned context, TOC, and suggestions 3. Create summary page using doc_write with source=<source_path> 4. Create/update concept pages with [[wikilinks]] 5. Periodically run wiki_lint to detect source-stale pages. It is categorised as a Execute tool in the MinerU Document Explorer MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MinerU Document Explorer MCP server in PolicyLayer and add a rule for wiki_ingest: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches MinerU Document Explorer. Nothing to install.
wiki_ingest is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the wiki_ingest rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for wiki_ingest. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
wiki_ingest is provided by the MinerU Document Explorer MCP server (opendatalab/mineru-document-explorer). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 15 MinerU Document Explorer tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
15 MinerU Document Explorer tools catalogued and risk-classified — across an index of 42,500+ MCP servers.