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 AnmarHani/SwarmVault --skill swarm-migrategit clone --depth 1 https://github.com/AnmarHani/SwarmVaultWrote 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/anmarhani/swarmvault/swarm-migrate)<a href="https://agentmods.dev/skills/anmarhani/swarmvault/swarm-migrate"><img src="https://agentmods.dev/badge/skills/anmarhani/swarmvault/swarm-migrate.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 | $0.00064 | $0.00623 |
| Opus 5 | $0.00032 | $0.00311 |
| Sonnet 5 | $0.00013 | $0.00125 |
| Haiku 4.5 | $0.00006 | $0.00062 |
Grade A, and why
swarm-migrate 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
swarm-migrate — existing-project migration
Three tiers per project; each deeper tier is opt-in with its cost stated up front. Idempotent: re-running repairs, never duplicates.
1. Discover
List candidates: swarmvault.py register --import-claude covers Claude Code users'
~/.claude.json in one shot; otherwise the user names paths. Ask WHICH projects to
migrate (multi-select) and, per project, the isolation flag.
2. Immediate tier (cheap, always)
Per chosen project: register + marker → first sync (mirrors existing
CLAUDE.md/AGENTS.md/README/docs and any existing memory; builds the MOC) → wire the
platforms they name with setup/wire (as swarm-init step 3; one command, no manual file
edits). The project is now queryable.
3. Digest tier (token-costing, opt-in)
State the rough cost first. Analyze the repo → 10 Projects/<P>/<P> Digest.md
(architecture, conventions, entry points — not generated, so sync preserves it) + seed
memories for footguns found in code comments/READMEs/issue notes.
4. SDLC adoption tier (most token-costing, opt-in) — place it IN the flow
Mine first, ask second (swarm-spec's brownfield mode does the work):
- Sources: README, docs/, existing plan/TODO files, issue exports, test suites, the code.
- Produce
docs/SRS.md+ feature specs markedstatus: mined-draft, every statement with provenance (file:line). - Detect the de-facto phase — e.g. "14 of 20 mined features implemented, no design doc" — and write flow-state accordingly.
- Gap interview via the question queue: only gaps, conflicts, low-confidence items. Never re-ask what the repo answers.
- Remaining work → tickets. From here
/swarm-flowresumes the project mid-phase like any SwarmVault project.
Close
Ask once (if unset): vault as default for new projects? Show the security disclaimer (swarm-init step 4) if this user hasn't seen it. Huge monorepo → scope the deep tiers to user-chosen subpaths. A project containing its own Obsidian vault → exclude that folder from doc mirroring (no vault-in-vault).
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 · 54 lines · 64 tokens per session scan A cde64f9f24b6
swarm-migrate is a skill published in the GitHub repository AnmarHani/SwarmVault (5 stars, last pushed 1mo ago), licensed MIT. It adds 64 tokens to every session and 623 once invoked, about $0.0003 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
knowledge-curation
Context priming before work (bd prime) and self-reflection after completion to extract patterns, gotchas, and decisions into the knowledge base.
planner
Use when the user asks to create, write, or review a docs/plans plan.
repo-bootstrap
Use when the user asks to bootstrap a new repo or project.
anvien-cli
Use when the user needs to run Anvien CLI commands like analyze/index a repo, check status, clean the index, generate a wiki, or list indexed repos. Examples: "Index this repo", "Reanalyze the codebase", "Generate a wiki".
agent-swarm-issue
Agent skill for swarm-issue - invoke with $agent-swarm-issue.
agent-sync-coordinator
Agent skill for sync-coordinator - invoke with $agent-sync-coordinator.