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/songzhuozhu/obsidian-llm-wikiWrote 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/songzhuozhu/obsidian-llm-wiki/ingest_raw)<a href="https://agentmods.dev/commands/songzhuozhu/obsidian-llm-wiki/ingest_raw"><img src="https://agentmods.dev/badge/commands/songzhuozhu/obsidian-llm-wiki/ingest_raw/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/songzhuozhu/obsidian-llm-wiki/ingest_raw"><img src="https://agentmods.dev/badge/commands/songzhuozhu/obsidian-llm-wiki/ingest_raw.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 | $0.00032 | $0.00824 |
| Opus 5 | $0.00016 | $0.00412 |
| Sonnet 5 | $0.00006 | $0.00165 |
| Haiku 4.5 | $0.00003 | $0.00082 |
Grade A, and why
ingest_raw 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 12d 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
Command: ingest_raw
You are the knowledge-maintenance agent for inspool-wiki-en/. Follow inspool-wiki-en/AGENTS.md strictly.
Task for this run: process the next ingest unit in inspool-wiki-en/raw/unprocessed/ and integrate it into the wiki.
Important: this run performs ingest only. Do not archive raw files yet. They must remain in pending_review until the user confirms the result.
Execute in this order:
- Read
inspool-wiki-en/AGENTS.md. - Review
inspool-wiki-en/raw/index.md,inspool-wiki-en/wiki/index.md, andinspool-wiki-en/wiki/log.md. - Scan
inspool-wiki-en/raw/unprocessed/. - Select the next ingest unit using this priority:
- files in the same subfolder
- files sharing the same
ingest_group - otherwise, a single file
- If multiple ingest units are possible, choose the oldest queued batch; if that is unclear, choose the first in lexical order.
- If the selected unit is a batch, verify that the files are actually related. If the batch is too large, split it into smaller batches of 2 to 5 files.
- Extract the core information from that source or source batch.
- Create or update the relevant source summary pages in
inspool-wiki-en/wiki/sources/and treat them as evidence nodes. - Create or update the related entity, concept, and synthesis pages.
- Add necessary
[[wiki links]], and prefer linking key claims to[[sources/...]]or[[sources/...#some-section]]. - Maintain frontmatter relationship fields on core pages such as
supports,contradicts,related_entities,related_concepts, andrelated_sources. - Update
inspool-wiki-en/wiki/index.md. - Append an
ingestlog entry toinspool-wiki-en/wiki/log.md. - Update workflow fields for every raw file in the ingest unit, for example:
processing_status: pending_reviewreviewed_at:processed_at:processed_into: <corresponding source page path>
- Update
inspool-wiki-en/raw/index.md. - Tell the user explicitly that this batch is now in
pending_review.
Constraints:
- the body content under
raw/is read-only by default - attach source-page references to important conclusions whenever possible
- prefer local source pages over external URLs as primary graph edges
- remain compatible with user-created subfolders under
wiki/concepts/,wiki/entities/, andwiki/synthesis/ - mark uncertain material as "to be confirmed" or an open question
- if a new source conflicts with older pages, record the conflict explicitly instead of silently overwriting it
- do not stop after creating a single summary page; maintain the related pages together
- raw files may be marked
pending_reviewonly afterwiki/index.md,wiki/log.md, andraw/index.mdare all updated successfully - process one ingest unit at a time; if it is a batch, keep it small and coherent
- do not move raw files to
processed/without user approval
If inspool-wiki-en/raw/unprocessed/ is empty:
- say clearly that there is nothing to process
- do not create empty logs or pages
- stop there
Output requirements:
- summarize which ingest unit you selected and which pages you plan to update
- summarize what was added or updated
- state clearly which raw files are now in
pending_review - call out insufficient evidence or naming conflicts if you find them
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.
- 12d ago First seen · 65 lines · 32 tokens per session scan A ca160fdffdbb
ingest_raw is a command published in the GitHub repository songzhuozhu/obsidian-llm-wiki (121 stars, last pushed 4mo ago), licensed MIT. It adds 32 tokens to every session and 824 once invoked, about $0.0002 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
brand-generate
Generate an on-brand document from a saved Brand Profile.
prose-review-docs
Reviews markdown documentation files or directories against the active prose style rules — the plugin's vendor-neutral baseline plus your organization's overlay when one is configured. Accepts a file path or directory (recursive). Runs prose-style-checker and optionally Vale. Supports --fix to auto-apply safe…
cti-report
Render case deliverables — relationship graph (PNG/SVG/Mermaid) and a polished PDF/DOCX assessment. Usage: /cti-report [--graph|--pdf].
ppt-image2-editable-rebuild
Rebuild image2 or imagegen reference slides as editable PowerPoint decks.
render-figures
Compile all .tex and .typ figure files in a directory.
harness-onboarding
Generate a human-readable onboarding document from HARNESS.md, AGENTS.md, and REFLECTIONLOG.md — a friendly guide for new team members.