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 tonydzi/second-brain-starter-kit --skill release-slicegit 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/release-slice)<a href="https://agentmods.dev/skills/tonydzi/second-brain-starter-kit/release-slice"><img src="https://agentmods.dev/badge/skills/tonydzi/second-brain-starter-kit/release-slice/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/tonydzi/second-brain-starter-kit/release-slice"><img src="https://agentmods.dev/badge/skills/tonydzi/second-brain-starter-kit/release-slice.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 | $0.00076 | $0.01585 |
| Opus 5 | $0.00038 | $0.00792 |
| Sonnet 5 | $0.00015 | $0.00317 |
| Haiku 4.5 | $0.00008 | $0.00159 |
Grade A, and why
release-slice 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 yesterday.
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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RELEASE-SLICE - the pipeline "a slice of the system → a public repo → a content wave" (the "free school" movement)
CONTEXT (canon = vault decision-open-second-brain-free-education-go-2026-07-02):
- We give away the skeleton and the lessons for FREE, and NEVER the contents or the data. MIT. We teach, we don't sell.
- The roadmap of pains =
%WORKDIR%\public-repos\claude-bible\ROADMAP.md(a public promise: releases on Mon + Thu). - GitHub pushes are autonomous (the owner, 2026-07-03: "POST it all yourself - I TRUST you"); Telegram/Facebook/channels keep their own gates.
- The repo family: every new repo declares its kinship (a link to claude-bible = the family map) + a FOR-ROBOTS.md.
STEPS (in order, skip none):
-
ANTI-DUPLICATE + RECALL: check whether a parallel session is already doing (or has done) this slice (grep the staging area
public-repos\,priority.json, recent retros, the bus). The fleet is active: someone else's work = join it, don't duplicate it. -
INVENTORY THE SLICE: what goes in (scripts from
_imports/~/.claude/scripts, the pattern write-ups, the pitfalls from memory). Estimate "how much personal material is inside" - that sets the depth of the scrubbing. Read the implementation; don't retell it from memory. -
STAGING: assemble the repo in
%WORKDIR%\public-repos\<name>\(NOT inside the vault). The minimum: README (the pain → the mechanics → a 5-minute quickstart → Versioning → Who made this + the WhatsApp CTA +1 341 222 9178 + the star ask "we need the first 10") · docs/ (the spec) · reference/ or templates/ (sanitized code/templates) · FOR-ROBOTS.md (the alpha, ranked + how to apply it) · LICENSE (MIT, Anton Dzyatkovsky) · CHANGELOG.md · devlog/. -
SANITIZE (by class, not spot fixes): real chat ids → placeholders; hostnames → roles (hub, laptop-1); keys only from env; no absolute paths like E:\ or C:; team and lead names → removed; the "scars" (bug war stories) stay - they are the value, but anonymized.
-
⛔ HARD GATE - LEAK-SCAN:
python $IMPORTS_ROOT/leak_scan.py <staging-dir>→ exit 0 (CLEAN) is mandatory; an INFO on the authorized CTA is normal; a FAIL gets fixed, never bypassed. The scanner normalizes whitespace and dashes - an ad-hoc grep is FORBIDDEN as a substitute. Plus a read-through with your own eyes.
5b. ⛔ THE MEANING GATE (owner's order, 2026-07-04): we publish PATTERNS, not live internals - no machine topology, no approval channels or their mechanics, no live control surfaces; publication LAGS behind production (what goes out is battle-worn or already replaced, never today's live circuit). Security wording must be provable ("layers of controls / blast radius / what stays with the human"; ⛔ "we make agents safe", "secure by design", promises of outcomes). Canon: vault reglament-security-yazyk-i-granitsy-publikatsiy + memory security-claims-language.
-
WRITING HYGIENE: em/en dashes are banned in the prose, only the short hyphen (the owner's signal, 2026-06-19); no invented numbers or durations (P14); run /taste-check when in doubt.
-
GIT: a publicly safe commit identity (
Anton Dzyatkovsky <[email protected]>),git init -b main→gh repo create tonydzi/<name> --public --source . --push→ tagv0.1.0→git push origin main --tags. Update claude-bible: the ROADMAP (pain → ✅ shipped as [link]) + the CHANGELOG (a new version, what went in) → push. Verification:gh repo viewsays PUBLIC +gh api .../tags. -
THE CONTENT WAVE: launch /wow with this slice's angle (it handles the rest itself: the episode → the approval chat gate → after a ➕, teasers into the RU and EN chats, a paste-ready Facebook post, a line in the log). The EN longread goes into the repo (docs/); the RU longread is written by the assistant who owns that language.
-
THE TRAIL: a line in the publication log (what → where → the link), a mark in the canon decision (the slice checkbox), a vault reindex if you wrote notes (
brain_embed_update.py, with --force for canon).
CADENCE: tagged releases on Monday and Thursday; small commits every day as things are ready. The promise is public, and we keep it.
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.
- yesterday First seen · 59 lines · 76 tokens per session scan A 4b09f17e71c9
release-slice is a skill published in the GitHub repository tonydzi/second-brain-starter-kit (6 stars, last pushed 2d ago), licensed MIT. It adds 76 tokens to every session and 1,585 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 2026-09-09.
Other skills, from other repositories
release-rollback
A release and rollback checklist for code or scripts, covering readiness checks, release decisions, and recovery after an incident.
carpeos-release
Cut a CarpeOS public release with consistent SemVer, CHANGELOG, git tags, GitHub Releases, and npm publish for @innocarpe/carpeos. Use when the user asks to release, bump version, publish to npm, create a git tag, make a GitHub release, ship @innocarpe/carpeos, run scripts/release.mjs, or keep Claude Code / Codex CLI…
finishing-a-development-branch
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup.
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.