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 Sixz-AI/repo-knowledge --skill rk-searchgit clone --depth 1 https://github.com/Sixz-AI/repo-knowledgeWrote 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/sixz-ai/repo-knowledge/rk-search)<a href="https://agentmods.dev/skills/sixz-ai/repo-knowledge/rk-search"><img src="https://agentmods.dev/badge/skills/sixz-ai/repo-knowledge/rk-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/sixz-ai/repo-knowledge/rk-search"><img src="https://agentmods.dev/badge/skills/sixz-ai/repo-knowledge/rk-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 | $0.00064 | $0.01985 |
| Opus 5 | $0.00032 | $0.00992 |
| Sonnet 5 | $0.00013 | $0.00397 |
| Haiku 4.5 | $0.00006 | $0.00198 |
Grade A, and why
rk-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 11d 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 — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RK Search — Two-Phase Semantic Knowledge Base Search
Search the knowledge base using breadth-first scanning across all projects, then deep verification of the best candidate.
Input
- Search query (natural language, any language)
- Optional: project name (if user explicitly specifies which project to search)
Step 1: Check for User-Specified Project
If the user explicitly names a project (e.g., "search cef-bridge for Execute"):
- Skip Phase 1 (Breadth Scan) entirely
- Go directly to Step 3 with that single project
- This avoids unnecessary scanning when the user knows where to look
Otherwise, proceed to Phase 1.
Step 2: Phase 1 — Breadth Scan (All Projects)
Shallow-scan every knowledge base in parallel to find the best candidates across all projects.
2.1: Discover Projects
- Read
~/.repo-knowledge/_registry.mdto get registered projects - Check if
~/.repo-knowledge/_personal/_index.mdexists - Build the scan list:
- If
_personalexists AND is not already in the registry: prepend_personalto the project list (highest priority) - Add all registered projects in registry order
- If
- If scan list is empty → tell user: "No knowledge bases found. Use
/repo-knowledge:rk-create <git-url>to create one, or/repo-knowledge:rk-memoto save personal knowledge." - If scan list has exactly 1 entry → skip scoring, go directly to Step 3 with that single project
2.2: Read All Indexes in Parallel
Read ~/.repo-knowledge/{project}/_index.md for every project in the scan list simultaneously.
2.3: Score Candidates
For each entry in each project's index, compute a score:
| Match Type | Score | Condition |
|---|---|---|
| Alias exact match | 3 (HIGH) | An alias exactly equals the query (case-insensitive) |
| Alias partial match | 2 (MEDIUM) | An alias contains the query or query contains the alias |
| Semantic match | 1 (LOW-MEDIUM) | Entry description semantically relates to the query |
Scoring rules:
- A single entry gets the highest matching score only (no double-counting)
- Add a project priority bonus:
_personalentries get +0.5; registry projects get +0.3 / position (1st = +0.3, 2nd = +0.15, 3rd = +0.1, etc.) to break ties - Only entries with score > 0 are candidates
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.
- 11d ago First seen · 189 lines · 64 tokens per session scan A 33c33ccb9e0d
rk-search is a skill published in the GitHub repository Sixz-AI/repo-knowledge (2 stars, last pushed 4mo ago), licensed MIT. It adds 64 tokens to every session and 1,985 once invoked, about $0.0003 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
procoder
Work like a senior developer in a repository governed by procoder: run the commit gate before calling anything done, format and lint through the binary, and drive the spec, plan, todo, backlog, and sprint chain in .procoder/. Use this skill when the repository contains a .procoder/ directory or an AGENTS.md naming…
explain
Guided code tour of a file or subsystem this session touched — entry point, the load-bearing pieces, the edges, and what connects to it.
x-spec
A system-planning guide that turns a vague idea into a set of linked design documents. It defines goals, modules, interfaces, data, workflows, and ways to check the result.
digest
Generate a structured vibe-learn learning digest of the current coding session from .vibe-learn/session-log.jsonl.
learn
Explain recent vibe-learn session activity or answer a question about what was built, grounded in .vibe-learn/session-log.jsonl.
quiz
Check your understanding of the vibe-learn session — recall questions grounded in the session log, with results tracked across sessions in the knowledge ledger.