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/s1lv4/th0th/findgit clone --depth 1 https://github.com/S1LV4/th0thWhat 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 | $0.00011 | $0.00243 |
| Opus 5 | $0.00005 | $0.00121 |
| Sonnet 5 | $0.00002 | $0.00049 |
| Haiku 4.5 | $0.00001 | $0.00024 |
Grade A, and why
find 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
Run a semantic code search with th0th.
Query: $ARGUMENTS
Steps:
- Resolve the active project:
- Prefer the projectId of the cwd if it's already indexed (call
th0th_list_projectsand match by path basename). - If ambiguous, ask the user.
- Prefer the projectId of the cwd if it's already indexed (call
- Call
mcp__th0th__th0th_searchwithquery="$ARGUMENTS",projectId=<resolved>,limit=10. - Return a ranked list of hits:
filePath:lineStart-lineEnd — score — label. For the top 3 results, include 3-5 lines of the matched snippet. - If 0 results, check whether vector store has orphaned chunks (look at previous
/mapoutput or prompt the user to/indexwithforceReindex=true).
Keep the output scannable — no walls of text.
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 · 20 lines · 11 tokens per session scan A 6238bbde2baf
find is a command published in the GitHub repository S1LV4/th0th (135 stars, last pushed 2mo ago), licensed MIT. It adds 11 tokens to every session and 243 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
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.
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.).
proofrag
Evaluate a RAG/LLM app — generate a golden set, judge it, and produce a scorecard.