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.
git clone --depth 1 https://github.com/Galbaz1/video-research-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/commands/galbaz1/video-research-mcp/ingest)<a href="https://agentmods.dev/commands/galbaz1/video-research-mcp/ingest"><img src="https://agentmods.dev/badge/commands/galbaz1/video-research-mcp/ingest/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/commands/galbaz1/video-research-mcp/ingest"><img src="https://agentmods.dev/badge/commands/galbaz1/video-research-mcp/ingest.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 | $0.00010 | $0.00798 |
| Opus 5 | $0.00005 | $0.00399 |
| Sonnet 5 | $0.00002 | $0.00160 |
| Haiku 4.5 | $0.00001 | $0.00080 |
Grade A, and why
ingest 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ingest: $ARGUMENTS
Manually add entries to the Weaviate knowledge store.
Phase 0: Determine Intent
If $ARGUMENTS is empty, ask the user:
AskUserQuestion:
questions:
- question: "What do you want to add to the knowledge store?"
header: "Ingest"
multiSelect: false
options:
- label: "Add a finding or insight"
description: "Store a research finding, insight, or fact with evidence tier"
- label: "Add content analysis"
description: "Store an analysis of a document, URL, or text"
- label: "Add a concept"
description: "Store a concept with its definition and relationships"
- label: "Describe what to add"
description: "Free-form description — I'll pick the right collection"
Phase 1: Pick Collection
Map the user's intent to the right collection. Call knowledge_stats() first to verify Weaviate is available.
| Intent | Collection |
|---|---|
| Finding/insight | ResearchFindings |
| Content analysis | ContentAnalyses |
| Video note | VideoAnalyses |
| Concept | ConceptKnowledge |
| Relationship | RelationshipEdges |
| Meeting/call note | CallNotes |
If $ARGUMENTS includes a collection name, use it directly.
Always call knowledge_schema(collection="<name>") to get the exact property names and types. Do not guess property names — the schema is the source of truth.
Phase 2: Gather Properties
If the user provided text or a file path in $ARGUMENTS:
- Read the content
- Extract the relevant properties automatically
- Show the user what will be ingested and ask for confirmation
If no content provided, ask the user for the required properties interactively.
Phase 3: Ingest
Call knowledge_ingest(collection="<name>", properties={...}).
On success: report the object UUID and offer:
- "Search for it:
/gr:recall <topic>" - "Add another entry"
- "Add related concepts or relationships"
On error:
- Schema validation error: The error message lists allowed
name:typepairs. Callknowledge_schema(collection="<name>")for full details, then retry with correct properties. - Weaviate not configured: "Set up Weaviate first:
/skill weaviate-setup"
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 · 93 lines · 10 tokens per session scan A b4b647d63332
ingest is a command published in the GitHub repository Galbaz1/video-research-mcp (23 stars, last pushed 1mo ago), licensed MIT. It adds 10 tokens to every session and 798 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
dejavu-weaviate-rescue
Weaviate corruption rescue toolkit — scan, quarantine-segment, redo-class. Codifies the 2026-05-14 LSM-segment rescue playbook.
vector.batch_exists
Check existence for multiple vector keys.
vector.batch_upsert
Upsert multiple vectors in one itemwise batch.
vector.collection.stats
Read facts for one vector collection.
ai-rag-setup
Set up RAG (Retrieval-Augmented Generation) system.
rag-pipeline-gen
Generate complete RAG pipeline with embeddings, vector DB, and retrieval.