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 skills/tonydzi/second-brain-starter-kit/content-minenpx skills add tonydzi/second-brain-starter-kit --skill content-minegit clone --depth 1 https://github.com/tonydzi/second-brain-starter-kitWrote 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/tonydzi/second-brain-starter-kit/content-mine)<a href="https://agentmods.dev/skills/tonydzi/second-brain-starter-kit/content-mine"><img src="https://agentmods.dev/badge/skills/tonydzi/second-brain-starter-kit/content-mine.svg" alt="Measured on agentmods" 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 | $0.00082 | $0.01275 |
| Opus 5 | $0.00041 | $0.00638 |
| Sonnet 5 | $0.00016 | $0.00255 |
| Haiku 4.5 | $0.00008 | $0.00128 |
Grade A, and why
content-mine 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 6d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CONTENT_MINE — sessions → funnel drafts (draft-first, taste delegated to me)
GOAL: mine our Claude Code sessions for content-worthy MOMENTS and drop them as DRAFTS into the content-factory funnel. Publishing happens only on the operator's explicit "+" through /episode. Nothing goes out.
The operator delegated the taste call — "what deserves to be content" — to me (2026-07-08). I don't ask what to add; I decide.
PATHS:
- Engine:
$IMPORTS_ROOT/content-factory\content_miner.py(same family as alpha-extraction / intention_mine; reads the livevault_sessionspool). - Funnel (intake):
content-factory\triage\posts.jsonl(+posts.md),source_kind: session. Writes throughvoice_triage.py append(dedup/upsert). - Digest:
content-factory\miner\candidates\cand-latest.md. Ledger:miner\captured.jsonl.
STEPS:
- SCAN (deterministic, 0 tokens). Recent by default:
python content_miner.py mine --days 7. Backlog / the whole archive:mine --all --operator Anton --cap 150. Flags:--day YYYY-MM-DD,--operator all,--cap N. NoteOVER_THRESHOLDandOUT. - Empty (SESSIONS=0 / OVER_THRESHOLD=0) → say "no fresh moments worth capturing" and stop.
- READ the digest
cand-latest.md(orOUT). Each ### = a sessionsrc | date | machine | operator [PRIV] {score · tags}+ a headline + snippets ([U]=the operator, [A]=Claude). Look at the top ones (HOT→WARM); on large volumes, fan out Sonnet judges over slices (seeminer\slices\), they return JSON verdicts, you dedup and capture. - JUDGE BY TASTE: keep the genuinely worthy and DIFFERENT stories (built/shipped it · a war with a bug / a root cause · a human+AI, multi-machine or consensus technique · a sharp insight · a "wow"). Drop routine, thin, forked duplicates (similar headline → one story), purely private material. Positioning: non-coder + AI is the trump card; the goal is an offer from an LLM company plus an audience of builders.
- CAPTURE the worthy ones:
python content_miner.py capture --src <cc:xxxxxxxx> --title "<hook>" --note "<the gist>" --tier teaser|medium|longread|dev-log --angle "<angle>" --visibility public|personal|private. Tier: teaser=a hook; medium=the public diary post; longread=a narrative; dev-log=dry and technical. [PRIV] / anything that can't be scrubbed →--visibility personal. The engine blocks secret-looking tokens and dedups by src. - BRIDGE INTO THE CANON (the law "an event → first a beat → then a post", [[show-canon-single-source]]): for the 1-2 strongest public moments create a draft BEAT in
04-Projects\show-canon\beats-inbox\using the templatebeats\_TEMPLATE-beat.md(+authored_by: content-miner); never write intobeats\directly — the canon writer moves them. Older (pre-canon, before 2026-07-08) events get a backfill beat only when an episode is actually being assembled from them. - REPORT to the operator (ELI5): how many sessions were scanned → how many drafts went into the funnel (with tiers) → how many beat candidates landed in the canon inbox; what's next = /episode on a "+", the outbound queue =
registry\pub_registry.py(queue → next → posted, limits in limits.json). Showpython content_miner.py captured --limit N.
BOUNDARIES (keep it simple and repairable):
- Draft-first, HARD: drafts only; no auto-publish; nothing goes out.
- Secrets/private material/CRM/leads never become content (engine gate + [[credential-store]]).
- Don't build a parallel monolith — reuse content_miner + voice_triage + /episode.
- Grunt work (detection/classification) = Sonnet; the authored post text = Opus (and that happens in /episode, not here).
- This is the manual twin of the nightly
content-miner-nightly; the archive backlog was closed in one pass on 2026-07-09 (miner\debt-log.md).
About & contact
Built and battle-tested at Palo Alto AI Research Lab — a fleet of Claude Code machines running 24/7 as a second brain and synthetic cofounder. Every skill here survived real production use before publication.
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.
- 6d ago First seen · 51 lines · 82 tokens per session scan A 70d08fe7841c
content-mine is a skill published in the GitHub repository tonydzi/second-brain-starter-kit (5 stars, last pushed 6d ago), licensed MIT. It adds 82 tokens to every session and 1,275 once invoked, about $0.0004 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 skills, from other repositories
wiki-ingest
Ingest supplied source material into an Obsidian vault with provenance and claim tracking: pasted text, files staged in the selected vault's inbox or .raw archive, or explicitly approved URLs. Use for a single source or bounded batch, not for saving an assistant answer. Triggers: ingest, ingest this file, ingest this…
autoresearch
Run a bounded, source-grounded research loop, draft a cited dossier, and optionally propose a separately reviewed canonical vault merge. Use when the user wants autonomous or deep research that may access the public web. Triggers: /autoresearch, autoresearch, research this topic, deep dive into, investigate, find…
canvas
Create, inspect, and update Obsidian JSON Canvas boards with text, file, link, group, and edge nodes. Use for canvas status, canvas lists, visual maps, zones, spatial layouts, adding vault notes or media to a .canvas file, and requests such as create canvas, add to canvas, or put this on the canvas.
wiki-retrieve
Build and query a vault-local contextual BM25 retrieval index with optional multilingual Nomic cosine reranking; use for retrieve, hybrid retrieval, BM25, rerank, contextual retrieval, chunk search, vault search, semantic search, find relevant passages, or retrieval diagnostics. Derived caches stay under .vault-meta…
wiki
Initialize, adopt, and route work for a separate Obsidian knowledge vault through the portable claude-obsidian core. Use for vault setup, scaffolding, workspace selection, cross-project configuration, or choosing the correct wiki sub-skill. Triggers: /wiki, set up wiki, scaffold vault, create knowledge base, adopt…
defuddle
Plan and, with explicit network consent, use an optional external Defuddle cleaner to extract article-like HTTPS pages as Markdown. Use for defuddle, clean this URL, strip page clutter, readable Markdown from a web page, or preparing a web source for later wiki ingestion.