New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Interlinked Semantic Index

by QuentinCody in QuentinCody/interlinked-cli

Install and operate Interlinked's optional local semantic function index.

MITAuto-check passedAI & LLM Engineering

Install Interlinked Semantic Index

skills CLI
$ npx skills add QuentinCody/interlinked-cli --skill interlinked-semantic-index -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install QuentinCody/interlinked-cli interlinked-semantic-index --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/QuentinCody/interlinked-cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/interlinked-semantic-index .claude/skills/interlinked-semantic-index && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
interlinked-semantic-index
GitHub stars
178
Token cost
~1.7k tokens
SKILL.md length
721 words
Files
2
Skills in repo
12
Repo updated
First seen
Licence
MIT

At a glance

Install and operate Interlinked's optional local semantic function index.

  • Works in 6 steps: Run interlinked semantic status and… → For model-missing, run semantic models;… → For runtime-missing, install/configure… → …
  • Tasks that involve Embeddings
  • SKILL.md covers Command surface, Configuration and storage, Index and query contract and Troubleshooting
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Interlinked Semantic Index is an agent skill from QuentinCody/interlinked-cli. Install and operate Interlinked's optional local semantic function index. Load this for interlinked semantic models/install/index/status/search/similar, local embedding runtime setup, model or index corruption, stale/model-mismatch states, or questions about canonical tokens versus embedding-model tokens. The feature is local-only and experimental; it never controls code-edit gates.

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in AI & LLM Engineering, covering Embeddings. It works with llama.cpp. The repository describes itself as: The harness for your harness. Local hooks, taste enforcement, and developer observability for AI coding agents (Claude Code, Codex, Cursor, Copilot CLI). The licence is MIT.

When your agent uses it

  • Tasks that involve Embeddings

Example prompts

  • “/interlinked-semantic-index”

Requirements

  • Python 3

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Run interlinked semantic status and preserve its state/reason.
  2. For model-missing, run semantic models; only install after the operator has authorized the
  3. For runtime-missing, install/configure compatible local llama.cpp executables; do not add a
  4. For stale, run semantic index; the last generation is still readable meanwhile.
  5. For corrupt or model-mismatch, run semantic index --rebuild after the configured exact
  6. For measurement-mismatch, run semantic index with the configured model/runtime available;

What it can do on your machine

Read from SKILL.md and the folder at commit a2adc8a. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are bash and json).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Interlinked Semantic Index loads about 1.7k tokens when it runs. Until then it costs about 104 tokens; SKILL.md has 721 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~104
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from QuentinCody/interlinked-cli at commit a2adc8a, republished under its MIT licence (© QuentinCody). 721 words, ~1,735 tokens.

Download SKILL.mdSave it as .claude/skills/interlinked-semantic-index/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
interlinked-semantic-index
description
Install and operate Interlinked's optional local semantic function index. Load this for `interlinked semantic models/install/index/status/search/similar`, local embedding runtime setup, model or index corruption, stale/model-mismatch states, or questions about canonical tokens versus embedding-model tokens. The feature is local-only and experimental; it never controls code-edit gates.

interlinked-semantic-index — local function retrieval

Interlinked can embed complete functions into a repository-local vector index and search them by meaning or similarity. The v1 subsystem is experimental, explicit, and local-only: no source, query, vector, or model inference is sent to the Interlinked MCP Server or a cloud provider.

This is independent from the hard function-token gate. The gate uses the interlinked-code-v2 contract (parser-resolved JS/TS syntax, Python stdlib tokenization) and an inclusive 500-token ceiling, without an embedding model. The semantic index uses the active model's real tokenizer, records modelTokens, and syntax-chunks long inputs before weighted-centroid aggregation. Model context changes never redefine or bypass the hard cap. The semantic runtime's llama-tokenize path currently requires the installed GGUF artifact and local llama.cpp commands, even though tokenization itself does not run neural inference.

Use syntax tokens (lexical tokens), not AST-node counts, when explaining the hard cap or metrics score. canonicalTokens is the compatible field name for the versioned syntax count; modelTokens describes the selected embedding tokenizer. No universal conversion rate exists between these units. The metrics composite, behavioral evidence and per-edit coverage workflows do not need semantic models. Route those operations to interlinked-quality-gates; installing an embedding model cannot resolve missing test evidence.

Command surface

bash
interlinked semantic models [--json]
interlinked semantic install --model <alias> [--json]
interlinked semantic index [--rebuild] [--include-tests] [--cwd <path>] [--json]
interlinked semantic status [--cwd <path>] [--json]
interlinked semantic search <query> [--top <n>] [--language <id>] [--path <glob>] [--cwd <path>] [--json]
interlinked semantic similar <file> --line <n> [--top <n>] [--cwd <path>] [--json]

Use models first. install is the sole download-authorized operation: it prints the license, size, source, and cache target, streams from an allowlisted HTTPS registry, verifies the pinned byte count and SHA-256, then atomically promotes the artifact. It does not build an index.

The default experimental manifest is the Apache-2.0 nomic-embed-text-v1.5-q4@0188c9bf409793f810680a5a431e7b899c46104c GGUF artifact. Interlinked does not execute model-repository code. In addition to the downloaded weights, the machine must provide compatible llama-embedding and llama-tokenize commands from llama.cpp. A runtime or model failure disables semantic results only; source edits and the guard daemon keep working.

Configuration and storage

Team policy is .interlinked/semantic.json:

json
{
  "version": 1,
  "enabled": false,
  "model": "nomic-embed-text-v1.5-q4@0188c9bf409793f810680a5a431e7b899c46104c",
  "include_tests": false,
  "include": ["src/**"],
  "exclude": []
}

Machine topology is .interlinked/semantic.local.json and may contain only device (auto or cpu), non-negative threads, batch_size, idle_unload_ms, incremental_indexing, and optional local command names llama_embedding_command / llama_tokenize_command. Unknown keys, including remote URLs and credentials, are rejected. Zero selects runtime auto-tuning.

Weights use the platform user cache (override with INTERLINKED_MODEL_CACHE). Generations use .interlinked/index/functions/generations/, with .interlinked/index/functions/CURRENT naming the last atomically published complete generation. Local config and all index artifacts are gitignored and are not included in sync.

Show full SKILL.md (351 more words)Show less

Index and query contract

  • index scans confined, ignored-aware product source with exact function adapters. Tests are excluded unless team config or --include-tests enables them; generated/vendor/data paths stay excluded. --rebuild disables unchanged-input vector reuse.
  • Every generation binds the exact model/runtime fingerprint, input schema, canonical counter, chunk aggregation policy, dimension, hashes, and source census. An interrupted build leaves the previous CURRENT generation readable.
  • status distinguishes absent, building, current, stale, corrupt, model-mismatch, model-missing, runtime-missing, and measurement-mismatch.
  • New metadata includes tokenMeasurement with the contract, language adapters and parser versions. Legacy/mismatched measurement metadata cannot be queried as current counts; status reports measurement-mismatch and exits 1. Run semantic index to refresh it. Unchanged full inputs and model/runtime fingerprints can reuse vectors while canonical counts and provenance are recomputed. --rebuild remains the explicit no-reuse option.
  • search embeds the query locally and exact-scans cosine similarity. similar uses the stored vector of the innermost indexed function containing the requested line and excludes itself. Results sort by descending score, then file, line, and symbol.
  • A stale index remains queryable with an explicit warning. A corrupt or fingerprint-mismatched index is refused; Interlinked never returns a partial generation.
  • Scores are heuristic retrieval evidence and comparable only within the named fingerprint. Query text is not persisted.

The enabled flag reserves automatic idle/incremental indexing policy. Manual semantic commands remain explicit operator actions. Current experimental builds do not schedule model work in a blocking hook response and do not offer remote inference, hybrid ranking, or an LLM reranker.

Troubleshooting

  1. Run interlinked semantic status and preserve its state/reason.
  2. For model-missing, run semantic models; only install after the operator has authorized the displayed download.
  3. For runtime-missing, install/configure compatible local llama.cpp executables; do not add a cloud fallback.
  4. For stale, run semantic index; the last generation is still readable meanwhile.
  5. For corrupt or model-mismatch, run semantic index --rebuild after the configured exact model/runtime is available. Do not hand-edit vectors, metadata, CURRENT, or fingerprints.
  6. For measurement-mismatch, run semantic index with the configured model/runtime available; this refreshes measurement metadata without forcing compatible inputs to be re-embedded.

Related skill: interlinked-quality-gates owns the separate deterministic 500-token ratchet.

© QuentinCody, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/interlinked-semantic-index of QuentinCody/interlinked-cli.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit a2adc8a

Compare with similar skills

Interlinked Semantic Index next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Interlinked Semantic Index compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Interlinked Semantic Index this skillQuentinCody/interlinked-cli178—~1.7kAutomated safety check: PassMIT
Pi AgentK-Dense-AI/scientific-agent-skills48k1 repos~2.1kAutomated safety check: PassMIT
Local Modelsglebis/claude-skills391—~1.4kAutomated safety check: PassMIT
Chroma Vector DatabaseOrchestra-Research/AI-Research-SKILLs13k7 repos~2.3kAutomated safety check: PassMIT
SageMaker Serving Image Selectionhuggingface/skills11k1 repos~4.6kAutomated safety check: PassApache-2.0
Codebase Managementgiancarloerra/SocratiCode3.3k1 repos~1.8kAutomated safety check: PassAGPL-3.0

Similar skills

  • Pi Agent

    K-Dense-AI/scientific-agent-skills

    Builds with and operates Pi, the minimal terminal coding harness.

    48k GitHub starsUsed in 1 repo~2.1k tokens
    AI & LLM EngineeringAuto-check passed

  • Local Models

    glebis/claude-skills

    Run quick, offline, private LLM tasks on local models via llama.cpp, reusing models already downloaded by Ollama.

    391 GitHub stars~1.4k tokensUpdated 2 days ago
    AI & LLM EngineeringAuto-check passed

  • Chroma Vector Database

    Orchestra-Research/AI-Research-SKILLs

    Shows how to store documents and embeddings in Chroma, query them by similarity with metadata filters, and persist them to disk for RAG and semantic search projects.

    13k GitHub starsUsed in 7 repos~2.3k tokens
    AI & LLM EngineeringAuto-check passed

  • SageMaker Serving Image Selection

    huggingface/skills

    Official

    Chooses the right serving container and current image URI for deploying a Hugging Face model to a SageMaker endpoint, preferring Hugging Face images over generic ones.

    11k GitHub starsUsed in 1 repo~4.6k tokens
    AI & LLM EngineeringAuto-check passed

  • Codebase Management

    giancarloerra/SocratiCode

    Set up, index, and manage SocratiCode codebase indexing. An agent skill from giancarloerra/SocratiCode.

    3.3k GitHub starsUsed in 1 repo~1.8k tokens
    AI & LLM EngineeringAuto-check passed

  • CLIP Image-Text Matching

    Orchestra-Research/AI-Research-SKILLs

    Explains OpenAI's CLIP model for zero-shot image classification, image-text similarity, semantic image search and content moderation, with install steps and code patterns.

    13k GitHub starsUsed in 7 repos~1.7k tokens
    AI & LLM EngineeringAuto-check passed

More from QuentinCody/interlinked-cli

All 12 skills in this repo

  • Interlinked Cowork

    QuentinCody/interlinked-cli

    Package, test, and operate the experimental Interlinked Cowork plugin.

    178 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed

  • Interlinked

    QuentinCody/interlinked-cli

    Overview and router for the Interlinked CLI — a local guard, quality-enforcement, simplification-review, semantic-code-search, and observability layer for AI coding agents.

    178 GitHub stars~4.1k tokensUpdated yesterday
    Auto-check passed

  • Interlinked Coordination

    QuentinCody/interlinked-cli

    Coordinate with other agents/humans via the optional Interlinked MCP Server, and use local checkpoints & file reservations.

    178 GitHub stars~2.9k tokensUpdated yesterday
    Auto-check passed

  • Interlinked Observability

    QuentinCody/interlinked-cli

    Investigate agent activity and JSONL/gzip evidence. An agent skill from QuentinCody/interlinked-cli.

    178 GitHub stars~10k tokensUpdated yesterday
    Auto-check passed

  • Interlinked Spec Audit

    QuentinCody/interlinked-cli

    Keep prose specs and design docs honest against the code using Interlinked's spec-audit system.

    178 GitHub stars~3.3k tokensUpdated yesterday
    Auto-check passed

  • Interlinked Supply Chain

    QuentinCody/interlinked-cli

    Respond to blocked package installs and manage the Interlinked supply-chain allowlist.

    178 GitHub stars~2.8k tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Interlinked Semantic Index

What does Interlinked Semantic Index do?

Install and operate Interlinked's optional local semantic function index. Interlinked Semantic Index is an agent skill from QuentinCody/interlinked-cli. Install and operate Interlinked's optional local semantic function index.

When should I use Interlinked Semantic Index?

Interlinked Semantic Index fits situations like: tasks that involve Embeddings.

How do I install Interlinked Semantic Index in Claude Code?

Run `npx skills add QuentinCody/interlinked-cli --skill interlinked-semantic-index -a claude-code`. Or copy the skill folder (skills/interlinked-semantic-index in QuentinCody/interlinked-cli) into .claude/skills/interlinked-semantic-index in your project. Claude Code loads it when a task matches its description.

How do I install Interlinked Semantic Index in Codex?

Run `npx skills add QuentinCody/interlinked-cli --skill interlinked-semantic-index -a codex`. Or copy the skill folder (skills/interlinked-semantic-index in QuentinCody/interlinked-cli) into .agents/skills/interlinked-semantic-index in your project. Codex loads it when a task matches its description.

Can I use Interlinked Semantic Index in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add QuentinCody/interlinked-cli --skill interlinked-semantic-index -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/interlinked-semantic-index, .gemini/skills/interlinked-semantic-index, .github/skills/interlinked-semantic-index and .opencode/skills/interlinked-semantic-index in your project.

What does Interlinked Semantic Index need to run?

SKILL.md names no scripts, command-line tools or credentials: Interlinked Semantic Index is instructions for the agent only. Our summary lists: Python 3.

Does Interlinked Semantic Index access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Interlinked Semantic Index safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Interlinked Semantic Index use?

Interlinked Semantic Index is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Interlinked Semantic Index use?

About 1.7k tokens (SKILL.md is roughly 6.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Interlinked Semantic Index?

Skills that share tags, products or a category with Interlinked Semantic Index: Pi Agent (K-Dense-AI/scientific-agent-skills, 48k stars), Local Models (glebis/claude-skills, 391 stars), Chroma Vector Database (Orchestra-Research/AI-Research-SKILLs, 13k stars) and SageMaker Serving Image Selection (huggingface/skills, 11k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Interlinked Semantic Index?

QuentinCody (a GitHub user) maintains it in QuentinCody/interlinked-cli, which has 178 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on October 9, 2026.

Source: QuentinCody/interlinked-cli on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.