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 nickroci/ultan --skill ultan-searchgit clone --depth 1 https://github.com/nickroci/ultanWrote 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/nickroci/ultan/ultan-search)<a href="https://agentmods.dev/skills/nickroci/ultan/ultan-search"><img src="https://agentmods.dev/badge/skills/nickroci/ultan/ultan-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/nickroci/ultan/ultan-search"><img src="https://agentmods.dev/badge/skills/nickroci/ultan/ultan-search.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 | $0.00132 | $0.00629 |
| Opus 5.5 | $0.00053 | $0.00252 |
| Sonnet 5.5 | $0.00026 | $0.00126 |
| Haiku 4.5 | $0.00013 | $0.00063 |
Grade A, and why
ultan-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 1mo 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.
What it actually says
Ultan Search
Talks to the running agent-mem-daemon over its Unix socket and returns either a single entry (with browseable neighbours) or a ranked list of matches.
Usage
Invoke with one argument — the script auto-detects mode:
- Wikilink or path —
[[projects/agent-mem/concepts/README]],projects/agent-mem/concepts/README,global/conventions/use-uv.md, etc. → returns the entry's full content plus siblings, subfolders, and parent README excerpt so you can navigate to neighbours without a second call. - Free-text query —
optimizer pruning,tier 3 interrupts,prefer python over javascript, etc. → returns top-8 BM25-ranked entries with score + snippet.
Explicit prefixes win if auto-detection is wrong: path:<x> or query:<x>.
Running it
Run the search.py that sits in this skill's own directory (the directory
containing this SKILL.md — shown in the skill listing):
python "<this-skill-dir>/search.py" "<wikilink-or-query>"
Use Bash to run this — the script writes the result as markdown to stdout.
It is stdlib-only; any python3 works.
When NOT to use this skill
- You want a synthesised answer fusing multiple entries — use
/ultan-advisor <question>instead. The advisor runs Sonnet (deep search) + Opus (synthesis) to produce a single referenced answer; this skill is the raw-fetch primitive. - The priming snippet you already received in
additionalContextcovers the question — don't drill down redundantly. - You're trying to write to the library — use
/ultan <text>or let the curator capture it from conversation; this skill is read-only.
Daemon dependency
This skill requires the agent-mem daemon to be running. If the socket isn't
present at ${AGENT_MEM_HOME:-~/.agent-mem}/priming.sock, the script prints a
clear error. With the Ultan plugin installed the daemon lazy-starts on its own
— retry after a turn or two; from source, start it with uv run agent-mem-daemon -v.
What ships with it
1 file 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.
- 1mo ago First seen · 43 lines · 132 tokens per session scan A 0e943b4ae449
ultan-search is a skill published in the GitHub repository nickroci/ultan (4 stars, last pushed 3mo ago), licensed MIT. It adds 132 tokens to every session and 629 once invoked, about $0.0005 per session on Opus 5.5. 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
project-knowledge
Use when a task needs shared repository knowledge, project memory recall, durable documentation updates, knowledge routing, trust/safety checks, correction or consolidation of stale docs, crystallizing reusable lessons, or validation of the knowledge core. Use before non-trivial work that depends on project…
research-to-knowledge
Use when a task asks for deep research, current external information, source-backed comparison, claim verification, source quality assessment, adversarial-source screening, repository/tool/practice evaluation, or saving researched findings into the knowledge core. Do not use for routine local code lookup or project…
openlore
Query and publish to an OpenLore knowledge repository over SSH using ordinary shell commands. Use when a task needs project documentation, runbooks, shared team knowledge, or a place to publish findings.
openlore-housekeeping
Audit and maintain a shared OpenLore knowledge repository. Use on a schedule or on request to find stale docs, broken links, unreviewed inbox items, and missing skill coverage, then publish an audit report.
memex-best-practices
Zettelkasten best practices for building a high-quality knowledge graph.
memex-agentic-memory
A-MEM-inspired agentic memory workflow for structured knowledge capture.