Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/francescofioredev/accretanpx agentmods add skills/francescofioredev/accreta/accreta-setupWrote 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/francescofioredev/accreta/accreta-setup)<a href="https://agentmods.dev/skills/francescofioredev/accreta/accreta-setup"><img src="https://agentmods.dev/badge/skills/francescofioredev/accreta/accreta-setup/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/francescofioredev/accreta/accreta-setup"><img src="https://agentmods.dev/badge/skills/francescofioredev/accreta/accreta-setup.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 | $0.00100 | $0.03021 |
| Opus 5 | $0.00050 | $0.01510 |
| Sonnet 5 | $0.00020 | $0.00604 |
| Haiku 4.5 | $0.00010 | $0.00302 |
Grade A, and why
accreta-setup 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 today.
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 — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Setting up an accreta knowledge base
This procedure is distilled from doing it, not from planning it. The order below is the order that worked; the warnings are things that actually went wrong.
Where this file came from, and what it needs.
npx skills addput it here, from the accreta repository.metadata.requiresabove names the earliest accreta release that has every command below: if one of them does not exist, compare it withaccreta --versionrather than working around the gap. Refresh withnpx skills update accreta-setup— which overwrites this file without asking, so anything you want to keep belongs somewhere else.
Before anything: look at the corpus
The single decision that shapes everything else is which preset fits, and it is decided by what the sources are, not by what the user says they want.
| Corpus | Preset | Tell |
|---|---|---|
| source code, one or more repositories | codebase |
files with extensions, a git history, imports between modules |
| papers, reports, standards, specs | research |
prose documents that cite each other, no call graph |
| anything else | (none) | the default vocabulary — note, source, concept, decision, synthesis |
Do not guess from a repository name. Read a few files. A repository full of markdown
specifications is a research corpus that happens to live in git.
If the corpus is genuinely mixed, prefer the default vocabulary over forcing one preset — page types are configuration and can be extended later, but a wrong preset produces pages described in terms that do not fit and are tedious to rename.
1. Initialize
accreta init --preset research # or --preset codebase, or no flag
This writes accreta.config.yaml, knowledge/, sources/, and AGENTS.md — the
constitution the maintaining agent follows.
init will not overwrite an existing AGENTS.md. If the project already has one, init
says so and writes nothing. Compose the constitution separately and merge it by hand; do not
delete theirs.
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.
- today Changed · +87 lines 22bee28fd719
- 8d ago First seen · 190 lines · 100 tokens per session scan A 220bf9465225
accreta-setup is a skill published in the GitHub repository francescofioredev/accreta (1 stars, last pushed today), licensed MIT. It adds 100 tokens to every session and 3,021 once invoked, about $0.0005 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
ingest
Populate the gnosis-mcp knowledge base — from local files, git history, or a crawled website. Handles the full matrix of flags (--force, --prune, --wipe, --embed, --include-crawled) in one place.
setup
First-time setup wizard for Gnosis MCP. Install, init the database, ingest a docs folder, wire your editor — in that order.
tune
Find the chunk-size and retrieval config that maximizes quality on YOUR corpus. Sweeps chunk sizes, runs a golden-query set, reports nDCG / MRR / Hit@5. Use after first ingest or whenever your corpus changes shape significantly.
karpathy-llm-wiki
Use when building or maintaining a personal LLM-powered knowledge base. Triggers: ingesting sources into a wiki, querying wiki knowledge, linting wiki quality, 'add to wiki', 'what do I know about', or any mention of 'LLM wiki' or 'Karpathy wiki'.
memobase
A Russian-language skill for searching a local knowledge base built from documents and other sources such as web pages, YouTube, audio, and Obsidian notes.
wiki-query
A knowledge-base search skill that retrieves existing information from a local collection of documents and combines it into an answer. It can use exact matches, broader search, and links between related pages.