Skip to content
Agentmart
Submit a tool

okf:search

Search concepts in an OKF bundle by query, type, tags, or full-text.

Skill by siculo

IndexedUnlicense★ 2
Write a review

Chalk Labs

Need help deploying okf:search?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/siculo-okf-search

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Search concepts in an OKF bundle by query, type, tags, or full-text. Usage: /okf:search <query> [type:<Type>] [tags:<tag>,<tag>] [bundle:<path>] [in:title|description|body|all]

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/okf-search

What's inside

Skill name
okf:search
Description
Search concepts in an OKF bundle by query, type, tags, or full-text. Usage: /okf:search <query> [type:<Type>] [tags:<tag>,<tag>] [bundle:<path>] [in:title|description|body|all]
License
Unlicense
Instructions
99 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: only lowercase letters, digits and single hyphens allowed
  • name: "okf:search" does not match folder "okf-skill-search"

Files (1)

  • SKILL.md

SKILL.md (first lines)

You are searching an OKF (Open Knowledge Format) bundle for concepts matching the user's query.

## 1. Parse arguments

`$ARGUMENTS` may contain, in any order:
- A free-text query string (the main search terms).
- `type:<value>` — filter by exact or partial `type` match (case-insensitive).
- `tags:<tag1>,<tag2>` — filter to concepts that have ALL the listed tags.
- `bundle:<path>` — path to the bundle root (default: current working directory).
- `in:title` / `in:description` / `in:body` / `in:all` — scope of text search (default: `in:all`).

If `$ARGUMENTS` is empty, ask the user for a query before proceeding.

Examples:
- `/okf:search orders` — full-text search for "orders"
- `/okf:search revenue type:Metric` — find Metrics mentioning revenue
- `/okf:search tags:sales,orders` — all concepts tagged both "sales" and "orders"
- `/okf:search freshness in:body bundle:./my-bundle` — body-only search in a specific bundle

## 2. Locate the bundle

Use the `bundle:` argument if provided, otherwise the current working directory. Verify it is an OKF bundle by checking for at least one `.md` file with frontmatter.

## 3. Build the search index

Follow `okf:_okf-skill-scan-bundle`. Use the returned index as the search index. Do not index `index.md` or `log.md` as searchable concepts.

## 4. Score and rank results

For each concept, compute a relevance score based on:

| Match location              | Weight   |
|-----------------------------|----------|
| `title` exact match         | Highest  |
| `title` partial match       | High     |
| `description` match         | High     |
| `tags` match                | Medium   |
| `type` match                | Medium   |
| Body text match             | Lower    |

Apply filters first (`type:`, `tags:`, `in:`), then rank by score. Return all matches above a relevance threshold; if no matches exceed the threshold, return the top 5 regardless and note low confidence.

For semantic relevance (e.g., "revenue" matching a concept about "income"), use your language understanding — OKF search is agent-native and does not require exact string matching.

In two-pass mode: score all concepts on frontmatter fields first (Pass 1). Then, for concepts that pass type/tags/title/description filters and may have body matches, read their full body before computing final body-text scores (Pass 2).

## 5. Display results

Print results in this format:

```
OKF Search: "<query>"
Bundle: <path>  |  Scanned: <N> concepts
─────────────────────────────────────────
<rank>. [<title>](<bundle-relative path>)
   Type: <type>  |  Tags: <tags>
   <description>
   Match: <where the match was found and a short excerpt>

<rank>. [<title>](<bundle-relative path>)
   …

─────────────────────────────────────────
<N> result(s)  |  Filters: <active filters or "none">
Bundle mode: <full / two-pass (N concepts, threshold 50)>
Skill version : 1.0
```

If there are no results:
```
No results for "<query>" in <N> concepts.
Suggestions:
  - Try broader terms
  - Available types: <list all distinct type values in the bundle>
  - Available tags: <list all distinct tags in the bundle>
```

## 6. Follow-up

After displaying results, offer these actions the user can take:
- **Open a result**: "Type the number of a result to read the full concept."
- **Refine search**: "Add `type:X` or `tags:Y` to narrow results."
- **Related concepts**: if the user selects a result, show its cross-links as further exploration paths.

If the user selects a result by number, read and display the full concept file (frontmatter + body), formatted as readable markdown. Then offer to open related concepts reachable via its cross-links.

## 7. Write invocation log

Follow `okf:_okf-skill-write-log` with:
- `skill`: `okf:search`, `version`: `1.0`
- `bundle`: absolute path to the bundle root used for this search
- `outcome`: `success`
- `concepts_added`: `null`
- `concepts_updated`: `null`
- `concepts_split`: `null`
- `conflicts`: `null`
- `note`: `"<N> result(s) for query"` using the result count from step 5

From siculo/okf-skills, Unlicense.

More from siculo/okf-skills

Signals

Stars
2
Forks
0
Open issues
2
Last commit (this folder)
Jun 27, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Unlicense
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from siculo

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get okf:search running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying okf:search?

CallGet help
Cite this page

okf:search skill. Agent Mart. https://agentmart.dev/skills/siculo-okf-search (last updated 2026-10-24)