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/andresanemic/lore-plugin/transmute-lorenpx skills add andresanemic/lore-plugin --skill transmute-loregit clone --depth 1 https://github.com/andresanemic/lore-pluginWrote 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/andresanemic/lore-plugin/transmute-lore)<a href="https://agentmods.dev/skills/andresanemic/lore-plugin/transmute-lore"><img src="https://agentmods.dev/badge/skills/andresanemic/lore-plugin/transmute-lore.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 | $0.00226 | $0.03938 |
| Opus 5 | $0.00113 | $0.01969 |
| Sonnet 5 | $0.00045 | $0.00788 |
| Haiku 4.5 | $0.00023 | $0.00394 |
Grade A, and why
transmute-lore 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 4d 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 — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Transmute Lore
The criteria is already there. It is in a README nobody finishes, in a CLAUDE.md that grew by
accretion, in a comment that says don't touch this, it breaks hydration — written by people who
paid for it, sitting in shapes nothing can load. Nothing is missing. Nothing is reachable either.
This skill operates a project's body of criteria. Eight modes, one skill:
- ADD — the project never applied the Lore method (or has a rough/incomplete
lore/). Valuable criteria is trapped in non-distillable forms: long READMEs, an everything-mixed instruction contracts (CLAUDE.mdorAGENTS.md), comments, tacit knowledge. ADD extracts it and lifts it to the six-piece standard without losing it and without inventing what never existed. - CLEAN — the project already has the standard, but carries per-project copies of thematic
modules that duplicate what its area already owns. CLEAN removes those modules, keeps
identidad.md/principios.md/index.md, and rewritesindex.mdto point at the area's modules — but only after confirming that criteria already lives in the area (or reporting it if it does not). - TRANSLATE — the Lore exists but its language is mixed or wrong (e.g. artifacts generated in English when the user works in Spanish). TRANSLATE standardizes every Lore artifact into one target language — content and localizable filenames — preserving structure, link integrity, confidence markers and — above all — meaning: it is a translation, never a rewrite.
- UPGRADE — the Lore is in the standard and in active use, but was written against an older version of these skills. It is structurally fine and materially behind: it lacks the gates and distinctions the kit learned after it was written. UPGRADE arbitrates the existing Lore against the current version of the kit and raises it, without rewriting what is already earned.
- PRUNE — the Lore is in the standard, in active use, and nothing in it is wrong. It decayed by accumulating correct things until their sum no longer fits the deliverable. PRUNE is the only mode that asks does any of this need to be here? and counts apparatus against content. What it shrinks is the deliverable's surface, not necessarily the corpus — see the warning in Phase 4. It is not CLEAN: nothing here is a duplicate.
- MYCELIUM — the Lore is in the standard and nothing in it is wrong, and you are about to lean on it for real work. MYCELIUM walks the network and reports which clues nothing runs: no step of any procedure runs them. It writes nothing. It is the only mode that asks can this fire? instead of is this correct? — and the two are independent, which is why no reading finds it. Un Lore sin micelio es un herbario — countable, not fireable. Ese es el silencio dentro del silencio: a check can run and its result never be consumed — two separate joints — and it is why MYCELIUM blocks the next operation until every finding is written or explicitly declined.
- LEAVE — the project outgrows Lore or the person chooses to leave (H13). LEAVE removes governance (
<!-- lore:always-on -->) but keepslore/and routing as plainenrutamiento.md, leavingFASES.mdwithleave:marker soUPGRADEcan return. Project stays buildable without the kit. - CRYSTALLIZE — the Lore is healthy and must travel as a single Markdown derivative into a chat, an AI project, or a notebook such as NotebookLM. It resolves the live routing into one traceable snapshot while leaving every source untouched. The snapshot is extractable: unpacking it rebuilds a mini-root whose routing table resolves, so a third person's AI session can work as closely as possible to the crystallizer's. It does not replace the six live artifacts and is not the normal format for work inside an agent or IDE that can read the live tree.
What ships with it
10 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.
- 4d ago First seen · 188 lines · 226 tokens per session scan A 0df0b37b5fb8
transmute-lore is a skill published in the GitHub repository andresanemic/lore-plugin (6 stars, last pushed 3d ago), licensed MIT. It adds 226 tokens to every session and 3,938 once invoked, about $0.0011 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
init
Initialize a personal Trove, a local file-based knowledge index for Claude Code. Use when the user says start/set up/create my trove, or when they want to remember things but no trove exists yet. Scaffolds the trove directory, INDEX.md and entries/ folder at user scope (/.claude/trove, follows you everywhere) or…
remember
Capture a durable learning into the personal Trove. Use when the user says remember this, note this, save to trove, add to my trove, or when a session surfaces a decision, gotcha, convention/preference, or external reference worth keeping. Writes one atomic entry file and updates INDEX.md.
index
Index the repository into Trove memory layers (symbols, modules, subsystems, architecture). Use when the user asks to index the repo, build the trove index, refresh symbols, or re-index after large changes.
query
Progressive retrieval from the Trove index. Use when the user asks what Trove knows about a subsystem, where code lives, or wants context assembled before editing. Runs architecture -> subsystem -> module -> symbol retrieval with a context budget.
context-ingest
The single entry point for getting anything new into your context architecture. Takes any input (documents, URLs, YouTube, meeting notes, pasted content) and triages it: dump to inbox, park in planned, or integrate into active docs. WHEN TO USE: "Here's a document — integrate it into my context" "Save this for later"…
bcos-wiki
Wiki zone manager for BCOS — the universal long-form and cross-cutting content destination per plugin-storage-contract.md Rule 2. Stores runbooks, SOPs, decision logs, how-tos, glossaries, post-mortems, FAQs, meeting notes, and plugin cross-cutting context (charters, transcripts, research, customer-call notes). Two…