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 agentmods add commands/ramonclaudio/skills/contextgit clone --depth 1 https://github.com/ramonclaudio/skillsWhat 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 | $0.00023 | $0.00547 |
| Opus 5 | $0.00012 | $0.00273 |
| Sonnet 5 | $0.00005 | $0.00109 |
| Haiku 4.5 | $0.00002 | $0.00055 |
Grade A, and why
context 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 2d 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.
What it actually says
Route based on the first token in $ARGUMENTS:
list—qmd context list. Show all configured contexts grouped by collection.add [path] "<text>"—qmd context add $ARGUMENTS.- One-arg form (
add "text") uses cwd and auto-detects the collection. - Two-arg form takes an explicit path. Path formats:
qmd://collection/path— virtual path inside a collection (recommended)/— global context, applies to every collection.or relative path — auto-resolved from cwd- Absolute filesystem path — collection detected by longest-prefix match
- Collection names must match
[a-zA-Z0-9_-]+.
- One-arg form (
rm <path>(aliasremove) —qmd context rm $ARGUMENTS. Accepts both filesystem paths andqmd://URIs.
Context is hierarchical: when a search hits a path, qmd concatenates ALL matching prefixes (global → root → specific subdir), joined with \n\n. Adding context at multiple levels is additive, not overriding.
What context actually does: the qmd MCP server returns the document's full hierarchical context block in every search result's context field. Claude (and other LLM callers) sees the context alongside the snippet so it knows what kind of document it's reading. Context is purely descriptive metadata. It does NOT influence chunk selection, reranking, or scoring — those are driven by the query terms and the optional intent parameter, not by the YAML context: field.
So context's job is to TELL the model "this snippet comes from a Convex schema example, not a tutorial blog post." A short, specific context (one sentence) is better than a long one because the full string gets prepended to every result's snippet.
The /qmd:add SKILL writes the initial root context once at clone/register time. Use this command to update the root context or add deeper paths later (e.g. qmd://next.js/docs/api → "Stable App Router API reference").
If $ARGUMENTS is empty or the subcommand is unrecognized, print: usage: /qmd:context <list|add|rm> [args].
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.
- 2d ago First seen · 29 lines · 23 tokens per session scan A 67d2080a24d9
context is a command published in the GitHub repository ramonclaudio/skills (4 stars, last pushed 3mo ago), licensed MIT. It adds 23 tokens to every session and 547 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 2026-08-31.
Other commands, from other repositories
ai-pipeline
RAG/embedding pipeline scaffolding — delegates to ai-data-engineer agent.
agent-brain-index
Index documents for semantic search.
agent
Add an AI agent / RAG backend (@convex-dev/agent) to the Convex app.
ingest
Manually add knowledge to the Weaviate store.
import
Import CSV, JSON, JSONL, or PDF files into a Weaviate collection.
build-search-index
Build or refresh a vault's LOCAL BM25 search index (wiki-meta/search-index.json) — a deterministic, plugin-free search tier that works on every vault, including those without Smart Connections. Idempotent (fingerprint check → no rewrite). (Skill build-search-index handles natural-language triggers.).