context-engine-guide
Skill1 starsChanged 4 months agoApache-2.0View on GitHub
Use when the user asks about Context Engine itself — available MCP tools, how to do semantic code search, repo discovery, or which sub-skill fits a task. Examples: \"What Context Engine tools are available?\", \"How do I search the codebase with context-engine?\"
In this repository77 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- archifySkill
- banner-designSkill
- brandSkill
- chrome-devtoolsSkill
- code-reviewSkill
- computer-useSkill
- context-engine-debuggingSkill
- context-engine-exploringSkill
- context-engine-syncSkill
- debuggingSkill
- designSkill
- design-systemSkill
- docs-seekerSkill
- docxSkill
- pdfSkill
- pptxSkill
- xlsxSkill
- kubernetesSkill
- planningSkill
- playwrightSkill
- repomixSkill
- researchSkill
- slidesSkill
- ui-ux-pro-maxSkill
- vk-askSkill
- vk-bootstrap-autoSkill
- vk-brainstormSkill
- vk-cloneSkill
- vk-contentSkill
- vk-cook-routerSkill
- vk-cookSkill
- vk-debugSkill
- vk-design-from-visualSkill
- vk-designSkill
- vk-docsSkill
- vk-docs-summarizeSkill
- vk-fix-fastSkill
- vk-fix-hardSkill
- vk-fixSkill
- vk-fix-sourceSkill
- vk-fix-typesSkill
- vk-fix-uiSkill
- vk-git-cmSkill
- vk-git-cpSkill
- vk-git-mergeSkill
- vk-git-prSkill
- vk-git-tagSkill
- vk-goalSkill
- vk-helpSkill
- vk-hoh-devSkill
- vk-hoh-planSkill
- vk-hoh-qaSkill
- vk-hoh-runSkill
- vk-integrate-polarSkill
- vk-integrate-sepaySkill
- vk-journalSkill
- vk-learnSkill
- vk-plan-archiveSkill
- vk-plan-fastSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 70
Use these to navigate the call graph without a dedicated impact tool — follow [calls: …] outward and [callers: …] inward.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Context Engine Guide
Quick reference for the Context Engine MCP — semantic codebase search backed by embeddings + call-graph expansion (port of vibervn-context-engine). Use it as the FIRST CHOICE for code search instead of grepping/reading blindly.
Always Start Here
For any task involving code understanding, search, debugging, or sync:
list-repos— discover indexed repos, grab therepoId, check freshness (lastCommitSha,status,chunks)- Match your task to a sub-skill below and read that skill file
- Follow the skill's workflow and checklist
Every search tool requires a
repoId. There is no implicit "current repo" — always resolve it vialist-reposfirst.
If
lastCommitSha≠ your localgit rev-parse HEAD(orstatus≠indexed), the index is stale → seecontext-engine-syncto push before trusting results.
Sub-Skills
| Task | Skill to read |
| Understand architecture / "How does X work?" | context-engine-exploring |
| Trace bugs / "Why is X failing?" | context-engine-debugging |
| Index / sync / re-index a (private) repo | context-engine-sync |
| Tools reference, repoId, freshness | context-engine-guide (this file) |
Tools Reference
| Tool | What it gives you |
list-repos | Discover indexed repos → repoId, name, status, lastCommitSha, chunks |
codebase-retrieval | Primary search. Semantic search across the whole repo → ranked snippets with file#Lstart-Lend + call-graph annotations ([callers:N], [calls: …]) |
file-retrieval | Targeted snippets from ONE known file — describe what you want instead of reading the whole file |
sync-status | Index status of an upload repo (status, chunkCount, lastCommitSha, lastIndexedAt, error) |
sync-init | Create an empty upload-backed repo (private codebase via upload, not git clone) |
sync-upload-url | Presigned PUT URLs to upload file contents to object storage |
sync-commit | Enqueue a FULL re-index after uploading all files |
sync-push | Enqueue an INCREMENTAL re-index (changed + deleted paths) |
The
sync-*tools are low-level. For day-to-day syncing prefer the bundledce-syncCLI (seecontext-engine-sync) which automates walking files, uploading, and installing a git post-commit hook.
Tool Selection: when to use what
| Goal | Use |
| "Where is the code that does X?" (unknown files) | codebase-retrieval |
| "How does feature Y work?" / high-level map | codebase-retrieval |
| Find a snippet inside a file you already know | file-retrieval |
| Find ALL occurrences of a known identifier | grep (exact-match is grep's job) |
| Read a specific known line range | Read |
codebase-retrieval does NOT know git history or version control — it reflects the last indexed snapshot only.
Result Annotations
codebase-retrieval snippets are tagged with call-graph metadata derived from the index:
core/apps/iam/src/app.ts#L1-87 [callers:1] [calls: Hono, createServiceApp, registerRoutes, …]#L1-87— file path + line range (open withReadto edit)[callers:N]/[callers: name +k more]— who calls this symbol[calls: …]— what this symbol calls
Use these to navigate the call graph without a dedicated impact tool — follow [calls: …] outward and [callers: …] inward.
Example
1. list-repos
→ repoId=be1730bd-… name=boltz-one status=indexed lastCommitSha=74302f9 chunks=1391
2. codebase-retrieval({repoId, information_request: "how the STS token exchange route is wired"})
→ core/apps/iam/src/api/sts.routes.ts#L1-27, core/apps/iam/src/app.ts#L33-48 …
3. Read core/apps/iam/src/api/sts.routes.ts:1-27 for exact content before editingDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.