Agent skill

Vera

by VeraTools in VeraTools/vera

Code search over the current repository. An agent skill from VeraTools/vera.

MITAuto-check passedAI & LLM Engineering

Install Vera

skills CLI
$ npx skills add VeraTools/vera --skill vera -a claude-code

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

GitHub CLI
$ gh skill install VeraTools/vera vera --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/VeraTools/vera.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/vera .claude/skills/vera && 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
vera
GitHub stars
118
Token cost
~1.2k tokens
SKILL.md length
592 words
Files
6 (incl. references)
Skills in repo
2
Repo updated
First seen
Licence
MIT

At a glance

Code search over the current repository. An agent skill from VeraTools/vera.

  • Works in 2 steps: Unfamiliar repository: vera overview… → Then the question's first lookup goes…
  • Tasks that involve Codebase onboarding
  • SKILL.md covers First search, Pick the tool, Search well and Recover from a miss, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Vera is an agent skill from VeraTools/vera. Code search over the current repository. Your first search in a repository goes through Vera, not grep, rg, or file reads. Run vera overview in an unfamiliar repository, vera search "<what the code does" for how or where something works, vera structural definitions <symbol or vera references <symbol when the question names a symbol, and vera grep "pattern" for exact text or regex. Fall back to grep only after a Vera call has missed.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 7 other files, including reference files (for example `agents/openai.yaml`, `references/install.md` and `references/mcp.md`).

It sits in AI & LLM Engineering, covering Codebase onboarding and Retrieval-augmented generation. It works with Model Context Protocol, Rust and ONNX. The repository describes itself as: Local code search combining BM25, vector similarity, and cross-encoder reranking. Parses 60+ languages with tree-sitter, runs entirely offline, and returns structured results… The licence is MIT.

When your agent uses it

  • Tasks that involve Codebase onboarding
  • Tasks that involve Retrieval-augmented generation

Example prompts

  • “<what the code does”
  • “pattern”
  • “/vera”

Workflow steps

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

  1. Unfamiliar repository: vera overview (languages, entry points, hotspots) instead of ls and README skimming.
  2. Then the question's first lookup goes through the table below. Do not "check with grep first".

What it can do on your machine

Read from SKILL.md and the folder at commit 6cb92d5. 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.

    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

Vera loads about 1.2k tokens when it runs, and up to ~4k if it reads all its reference files. Until then it costs about 114 tokens; SKILL.md has 592 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~114
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~4k

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 VeraTools/vera at commit 6cb92d5, republished under its MIT licence (© VeraTools). 592 words, ~1,188 tokens.

Download SKILL.mdSave it as .claude/skills/vera/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
vera
description
Code search over the current repository. Your first search in a repository goes through Vera, not grep, rg, or file reads. Run `vera overview` in an unfamiliar repository, `vera search "<what the code does>"` for how or where something works, `vera structural definitions <symbol>` or `vera references <symbol>` when the question names a symbol, and `vera grep "pattern"` for exact text or regex. Fall back to grep only after a Vera call has missed.

Vera

Ranked code search over an indexed repository. Results are markdown codeblocks: path:line_start-line_end symbol_type:symbol_name (split symbols render as name (part N) in that position, the bare name plus the part suffix), then the code.

The first search action in a task decides whether Vera gets used at all; later searches follow the first tool chosen. So:

  1. Unfamiliar repository: vera overview (languages, entry points, hotspots) instead of ls and README skimming.
  2. Then the question's first lookup goes through the table below. Do not "check with grep first".

Pick the tool

Question shapeDo this
How or where does something workvera search "env file loading decision"
The question names a function, class, or config keyvera structural definitions make_config, then vera references make_config for callers
Who calls or what is called by a symbolvera references make_config / vera references make_config --callees
Exact text or regexvera grep "before_request" (line numbers always shown; --limit <N> caps results)
Enumerate every route, env read, implementationvera structural routes / vera structural env / vera structural impls <symbol>
Documentation on a topicvera search "deploying behind proxy" --scope docs
Files changed in this branchadd --changed, --since <rev>, or --base <rev> to any of the above
Edit the same pattern in many files mechanicallyrg
Read a file you already know the path and lines ofRead it directly

vera references resolves split-symbol call sites, vera structural definitions finds split symbols by bare name and deduplicates to the earliest declaration, and vera dead-code deduplicates split parts by (symbol, file).

Search well

  • Search behavior, not nouns: "JWT expiry handling", not "auth" or "utils".
  • Pass several angles in one call: vera search "OAuth token refresh" "JWT expiry" "auth middleware".
  • Start broad with --compact (signatures only, fewer tokens), then narrow with --lang, --path, --type, --limit.
  • --path is relative to the repository root: --path src/flask, --path "tests/**/*.py". Absolute paths inside the repository are accepted and rewritten; paths outside it match nothing.
  • Add --intent "<goal>" when the query is vague but the goal is clear.
  • --deep expands the query and merges rankings; use it only after normal search misses.
Show full SKILL.md (241 more words)Show less

Recover from a miss

  • Top hit is a usage site, not the definition: vera structural definitions <symbol> with the name from the hit. Do not try more phrasings.
  • Hit shows a call but not the caller chain: vera references <symbol>.
  • Two searches returned the same region: stop searching and read that code.
  • The code is in a dependency (site-packages, uv or pip caches, vendored trees outside the index): use rg there, then return to Vera for repository code.

Treat hits as leads

  • A hit is a lead, not evidence. Verify behavior against the hit's code; open the file only for lines the hit did not include.
  • Cite path:line from code you actually read.
  • A stale-index warning does not invalidate hits. After editing files, run vera update . from any directory; it refreshes the repository root's index.

Recovery

SymptomFix
no index foundvera index . from the repository root, then rerun the search
no indexed file matches <path>Use a root-relative --path; stderr suggests a wildcard directory form when one applies
Stale results after editsvera update . (or vera watch .); works from any subdirectory
A file is missing from resultsvera explain-path path/to/file
Local model or ONNX errorvera doctor --probe, then references/troubleshooting.md
Missing local assetsvera repair
Install, API keys, backendsreferences/install.md
MCP serverreferences/mcp.md

References

  • references/install.md: install, setup, API and local config, .veraignore rules
  • references/query-patterns.md: more query examples and rg guidance
  • references/troubleshooting.md: common errors and fixes
  • references/mcp.md: optional MCP server usage

© VeraTools, 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 5 other files (references) in skills/vera of VeraTools/vera.

  • SKILL.md
  • agents/openai.yaml
  • references/install.md
  • references/mcp.md
  • references/query-patterns.md
  • references/troubleshooting.md

Open the folder on GitHubat commit 6cb92d5

Compare with similar skills

Vera 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.

Vera compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Vera this skillVeraTools/vera118—~1.2kAutomated safety check: PassMIT
tilth Code Reading CLIjahala/tilth352—~1.1kAutomated safety check: PassMIT
Agentsop Repo Mapagentsope/SkillAlchemy459—~3.9kAutomated safety check: PassMIT
Codeguaardvark/guaardvark2551 repos~559Automated safety check: PassMIT
Trusty Searchbobmatnyc/claude-mpm155—~1.5kAutomated safety check: PassApache-2.0
SynalinksSynaLinks/synalinks-skills907—~4.8kAutomated safety check: PassApache-2.0

Similar skills

  • Replaces grep, cat, find and ls with the tilth CLI, which returns AST-aware outlines, definitions, usages and callers across many languages in one call.

    352 GitHub stars~1.1k tokensUpdated 11 days ago
    DevelopmentAuto-check passed
  • Agentsop Repo Map

    agentsope/SkillAlchemy

    Symbol-level code context for LLM coder-agents: tree-sitter extracts symbols, PageRank ranks them over the cross-file reference graph, and the top class/function signatures are fed to the LLM as a…

    459 GitHub stars~3.9k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Code

    guaardvark/guaardvark

    Code intelligence over repositories Guaardvark has indexed: semantic code search, exact-text search, file listing, PageRank repository map, import dependency graph, reading one class or function…

    255 GitHub starsUsed in 1 repo~559 tokens
    AI & LLM EngineeringAuto-check passed
  • Trusty Search

    bobmatnyc/claude-mpm

    Hybrid code search (BM25 + vector + KG) with RRF fusion. An agent skill from bobmatnyc/claude-mpm.

    155 GitHub stars~1.5k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Synalinks

    SynaLinks/synalinks-skills

    A skill your agent uses for anything involving the Synalinks neuro-symbolic LM framework (Keras-inspired): DataModel/Field/Input, JSON operators (+ & | ^ ~), synalinks.ops…

    907 GitHub stars~4.8k tokensUpdated 9 days ago
    AI & LLM EngineeringAuto-check passed
  • Agent Squad for TypeScript

    2FastLabs/agent-squad

    Guide to building Node.js and TypeScript apps on the agent-squad package: orchestrator, agent types, classifier routing, storage, retrievers and MCP tools.

    7.8k GitHub stars~4.3k tokensUpdated today
    AI & LLM EngineeringAuto-check passed

More from VeraTools/vera

  • Vera

    VeraTools/vera

    Code search over the current repository. An agent skill from VeraTools/vera.

    118 GitHub stars~825 tokensUpdated 8 days ago
    Auto-check passed

Questions about Vera

What does Vera do?

Code search over the current repository. An agent skill from VeraTools/vera. Vera is an agent skill from VeraTools/vera. Code search over the current repository.

When should I use Vera?

Vera fits situations like: tasks that involve Codebase onboarding; tasks that involve Retrieval-augmented generation.

How do I install Vera in Claude Code?

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

How do I install Vera in Codex?

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

Can I use Vera 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 VeraTools/vera --skill vera -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/vera, .gemini/skills/vera, .github/skills/vera and .opencode/skills/vera in your project.

What does Vera need to run?

SKILL.md names no scripts, command-line tools or credentials: Vera is instructions for the agent only.

Does Vera 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 Vera 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 Vera use?

Vera 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 Vera use?

About 1.2k tokens (SKILL.md is roughly 4.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2.9k tokens, read only when the agent opens those files.

What are the alternatives to Vera?

Skills that share tags, products or a category with Vera: tilth Code Reading CLI (jahala/tilth, 352 stars), Agentsop Repo Map (agentsope/SkillAlchemy, 459 stars), Code (guaardvark/guaardvark, 255 stars) and Trusty Search (bobmatnyc/claude-mpm, 155 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Vera?

VeraTools (a GitHub organization) maintains it in VeraTools/vera, which has 118 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 30, 2026.

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