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 kasssandr/archilles --skill archilles-searchgit clone --depth 1 https://github.com/kasssandr/archillesWrote 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/kasssandr/archilles/archilles-search)<a href="https://agentmods.dev/skills/kasssandr/archilles/archilles-search"><img src="https://agentmods.dev/badge/skills/kasssandr/archilles/archilles-search/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/kasssandr/archilles/archilles-search"><img src="https://agentmods.dev/badge/skills/kasssandr/archilles/archilles-search.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00071 | $0.01443 |
| Opus 5 | $0.00036 | $0.00722 |
| Sonnet 5 | $0.00014 | $0.00289 |
| Haiku 4.5 | $0.00007 | $0.00144 |
Grade A, and why
archilles-search 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 12d 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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Searching an ARCHILLES library
ARCHILLES indexes a personal research library — Calibre, Zotero, plain folders, Obsidian vaults — and retrieves passages with the page number they were printed on. The MCP tools expose retrieval only. You do the reading and the writing, and you are responsible for every claim you attribute to a source.
The tool schemas tell you what the parameters are. This skill tells you what they are for.
Pick the tool before you pick the query
| The user wants | Use |
|---|---|
| An answer argued from the library | search_books_with_citations |
| Everything by one author | list_books_by_author |
| Their own highlights and notes | search_annotations |
| Which tags exist, before filtering | list_tags |
| A bibliography file (BibTeX, RIS, …) | export_bibliography |
| Annotations from one specific file | get_book_annotations |
| The index brought up to date | watchdog_scan |
The common mistake is reaching for search_books_with_citations to answer
"what does the library have by Mason?". Vector search finds passages, not
holdings. A short article or a book chapter may contribute two chunks to an
index of millions and never surface. list_books_by_author queries Calibre's
metadata directly and gives a complete list. Use it whenever the question is
about holdings rather than content.
search_books_with_citations returns a prompt, not an answer
The tool hands back a system prompt and a user prompt containing the retrieved
excerpts as <document> blocks with ids doc_1, doc_2, … . Work under those
instructions:
- Answer only from the excerpts. If they do not carry the answer, say so. Do not fill the gap from your own knowledge of the subject — the whole point of the system is that the user can check every sentence against their own shelf.
- Cite each factual claim inline as
[doc_1], several as[doc_1, doc_3]. - Answer in the user's language while keeping scholarly terminology in the original.
Then convert the markers
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.
- 12d ago First seen · 138 lines · 71 tokens per session scan A 4d6e1f540efb
archilles-search is a skill published in the GitHub repository kasssandr/archilles (8 stars, last pushed yesterday), licensed MIT. It adds 71 tokens to every session and 1,443 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 2026-08-31.
Other skills, from other repositories
shared-memory
Search, save, and query a three-tier semantic memory shared across all AI agents on your workstation. Use before starting any task (search first for prior context) and after completing significant work (save findings with entities for consolidation). Supports savedecision for full PROV-O provenance and…
kami-deck
A lab-meeting deck on gut-microbiome links to sleep quality — the design, the results, the caveats, and the next experiment. Built as a decision-grade academic research deck for lab group, PI.
code-review-excellence
This skill should be used when the user asks to review a diff or pull request, write review comments, audit code quality, establish review standards, or improve how a team performs code review.
architecture-design
Use only when creating new registrable ML components that require Factory or Registry patterns.
git-workflow
This skill should be used when the user asks to "create git commit", "manage branches", "follow git workflow", "use Conventional Commits", "handle merge conflicts", or asks about git branching strategies, version control best practices, pull request workflows. Provides comprehensive Git workflow guidance for team…
results-analysis
This skill should be used when the user asks to "analyze experimental results", "run strict statistical analysis", "compare model performance", "generate scientific figures", "check significance", "do ablation analysis", or mentions interpreting experiment data with rigorous statistics and visualization. It focuses on…