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/rockyhong/super-bootstrap/drainnpx skills add RockyHong/super-bootstrap --skill draingit clone --depth 1 https://github.com/RockyHong/super-bootstrapWrote 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/rockyhong/super-bootstrap/drain)<a href="https://agentmods.dev/skills/rockyhong/super-bootstrap/drain"><img src="https://agentmods.dev/badge/skills/rockyhong/super-bootstrap/drain.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 | $0.00200 | $0.03178 |
| Opus 5 | $0.00100 | $0.01589 |
| Sonnet 5 | $0.00040 | $0.00636 |
| Haiku 4.5 | $0.00020 | $0.00318 |
Grade B, and why
drain scanned grade B with 1 finding 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 5d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Never `Read` a path under `.claude/worktrees/{id}/` — a worktree-internal Read re-injects that worktree's nested CLAUDE.md + rules per file and blows the context budget. Use the read-around paths (`git show {branch}:<pat How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
drain — Parallel-Worktree Auto-Drain
One wave per invocation over the board. Scan → Cloud-gate → wave-select → confirm → spawn one isolated subprocess per item → each halts at a user wall. The orchestrator (gateway) holds no in-head state; every tick re-reads from files. Capacity ceiling = how many halts the user can resolve, not machine throughput.
Consumer contract: assumes the super-bootstrap harness shape — docs/work/{BUG,DEBT,GAP}-###.md card files, /super-bootstrap:merge, /super-bootstrap:commit. Not portable below that line. First run self-installs the worktree infra (§Pre-flight step 0).
Trigger: user types /super-bootstrap:drain. Never auto-fires.
Invariants
- One wave, one shot per invocation. No internal loop across waves, no
--all. Turn ends after the wave is dispatched. Next invocation cold-reads files and re-picks. - No auto-merge — ever. Each subprocess stops at a ready-to-merge state. The user confirms; the merge runs via
/super-bootstrap:merge(the destructive-git lane). Subprocesses are denied push/merge/rebase/branch-delete/worktree at the permission layer. - Admission-gate, not type-gate. Eligible = the item's next phase is drainable, across BUG/DEBT/GAP card types. When the scale module is wired the gate is next-phase venue ∈ {T, S} (
.claude/rules/venue-map.md); without it the gate falls back tointent == Cloud(cloud-safe). Either wayDevice/Discussand venue U/P defer — drain never spawns for them. A mislabel is fixed upstream (clarify the card, the shared criterion, or the venue map), never overridden here. - Stage-resume. Each item's brief enters the chain at its current stage (thread-state):
raw→ground,triaged→plan,aimed→plan,executing→execute,review→review. Each phase is named by the artifact it lands (assets/phase-loop.md §Phase → artifact). Committed upstream phases are inherited, not re-run. - Drain-till-wall; halts are outcomes. Drainable is a phase attribute, not a ticket attribute: admission scores only the next phase, and the session runs phase-to-phase to its first wall (typed
user|shape—assets/phase-loop.md §Walls), then halts. The wall's finding plus any committed earlier phases are progress, not waste; a resolved wall re-enters on the next invocation. - Wave member = no blocker. Orphans + chain-heads enter; chain-tails and conflicts defer to a later invocation. No forward projection — render the current wave only.
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 113 lines · 200 tokens per session scan B 5b1be9f8dc6c
drain is a skill published in the GitHub repository RockyHong/super-bootstrap (3 stars, last pushed today), licensed MIT. It adds 200 tokens to every session and 3,178 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
create-site
Creates a new Power Pages code site (SPA) using React, Angular, Vue, or Astro. Guides through the full process from initial concept to deployed site: requirements discovery, scaffolding, component planning, design, implementation, validation, and deployment. Use when the user wants to create, build, or scaffold a new…
add-cloud-flow
Integrates Power Automate cloud flows into a Power Pages site. Lists available flows, suggests relevant ones based on intent, identifies scenarios and web roles, creates metadata files, and generates client-side code to call flows. Handles both new flow registration and adding already-registered flows to additional…
configure-env-variables
Configures environment variables for Power Pages site settings to support ALM across environments. Creates environment variable definitions in Dataverse, guides the user through linking site settings to those variables via the Power Pages Management app, adds the variables to the solution, and generates a…
deploy-site
Deploys an existing Power Pages code site to a Power Pages environment using PAC CLI. Handles tooling verification, authentication, environment confirmation, building, and uploading. Use when the user wants to deploy, upload, or publish their code site.
add-teams
Adds Microsoft Teams connector to a Power Apps code app. Use when sending Teams messages, posting to channels, or integrating with Teams chat.
tidewave-integration
Tidewave MCP runtime tools — debugging, smoke testing, live state inspection, SQL queries, hex docs. Use when evaluating code in a running Phoenix app.