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 skills add thoreinstein/gemini-obsidian --skill wiki-ingestgit clone --depth 1 https://github.com/thoreinstein/gemini-obsidianWrote 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/skills/thoreinstein/gemini-obsidian/wiki-ingest)<a href="https://agentmods.dev/skills/thoreinstein/gemini-obsidian/wiki-ingest"><img src="https://agentmods.dev/badge/skills/thoreinstein/gemini-obsidian/wiki-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/skills/thoreinstein/gemini-obsidian/wiki-ingest"><img src="https://agentmods.dev/badge/skills/thoreinstein/gemini-obsidian/wiki-ingest.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00064 | $0.01207 |
| Opus 5.5 | $0.00026 | $0.00483 |
| Sonnet 5.5 | $0.00013 | $0.00241 |
| Haiku 4.5 | $0.00006 | $0.00121 |
Grade A, and why
wiki-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 1mo 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- wiki-ingest — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wiki Ingest
Distill external sources into vault-native notes following VAULT_CONVENTIONS, with provenance tracking and deduplication.
Workflow
1. Identify Sources
Accept one or more source paths from the user. Sources can be:
- Files on disk (markdown, PDF, text, transcripts)
- Vault-internal notes (Readwise highlights, working notes to promote)
- A directory to scan for ingestible files
For vault-internal sources, use obsidian_read_note. For external files, use the Read tool.
2. Check Manifest
Read .manifest.json from the vault root using Read tool.
For each source path, check if it exists in sources:
- Not present → new source, proceed to extract
- Present but source modified since
ingested_at→ re-ingest - Present and unchanged → skip (tell user "already ingested, use --force to re-process")
3. Extract Knowledge
Read the source content. Identify:
- Patterns — reusable approaches, techniques, or practices
- Traps — pitfalls, anti-patterns, things that went wrong
- Decisions — architectural or design choices with rationale
- Concepts — definitions, mental models, frameworks
For each extraction, classify its provenance:
- Extracted (unmarked) — directly stated in the source
- Inferred
^[inferred]— synthesized, generalized, or implied but not stated - Ambiguous
^[ambiguous]— source is unclear or contradicts other sources
4. Deduplicate
For each extracted concept, search the vault:
- Call
obsidian_rag_querywith the concept name/summary,limit: 5 - If a strong match exists, read the existing note with
obsidian_read_note - Decide: merge into existing note or create new note
Merge when: the existing note covers the same concept and the new source adds information. Create when: no existing note covers this concept, or the angle is sufficiently different.
5. Present Plan
Before creating or modifying any notes, present the plan to the user:
## Ingest Plan for: <source name>
### New Notes
- Engineering/Patterns/<name>.md — <one-line summary>
- Engineering/Traps/<name>.md — <one-line summary>
### Merge Into Existing
- Engineering/Patterns/<existing>.md — adding: <what's new>
### Skipped
- <concept> — already covered in [[existing-note]]
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.
- 1mo ago First seen · 135 lines · 64 tokens per session scan A 4dabf9b17cee
wiki-ingest is a skill published in the GitHub repository thoreinstein/gemini-obsidian (103 stars, last pushed 3d ago), licensed ISC. It adds 64 tokens to every session and 1,207 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
expert-librarian
An expert librarian that helps you organize, search, and synthesize your Readwise highlights and documents.
im-local-kb
A local knowledge-base manager for turning chat logs into organized Markdown records and sourced project documents. A knowledge base is a searchable collection of stored information.
oma-checkpoint
Record an immutable Session Trail checkpoint with source hashes and a compact resume pack.
oma-deep-init
Run deep repository initialization to create durable project-map context for long sessions.
oma-recall
Recall prior decisions and evidence with state-first search and bounded history fallback.
oma-resume
Inspect Session Trail continuity and the next action without restoring stale state or clearing locks.