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 aleksbuss/orchestra --skill graphifygit clone --depth 1 https://github.com/aleksbuss/orchestraWrote 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/aleksbuss/orchestra/graphify)<a href="https://agentmods.dev/skills/aleksbuss/orchestra/graphify"><img src="https://agentmods.dev/badge/skills/aleksbuss/orchestra/graphify/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/aleksbuss/orchestra/graphify"><img src="https://agentmods.dev/badge/skills/aleksbuss/orchestra/graphify.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 | $0.00077 | $0.02349 |
| Opus 5.5 | $0.00031 | $0.00940 |
| Sonnet 5.5 | $0.00015 | $0.00470 |
| Haiku 4.5 | $0.00008 | $0.00235 |
Grade A, and why
graphify 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 1mo 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Graphify — code navigation by graph, not by grep
Graphify parses a repository into a local knowledge graph and answers structural questions without
reading whole files into context. Code is parsed with tree-sitter — deterministic and offline. (Non-code
documents go through an LLM pass when a provider key is present; inside code_execution there is no
key, so what you get is the free AST path.)
This skill is a wrapper around an external CLI. It is not installed by default. Everything below
runs through code_execution. See docs/references/graphify-integration-adr.md
for why this is a skill and not a tool.
Decision — graphify first, grep second (read this before you touch find/grep)
You loaded this skill because a codebase question came up. The measured failure mode is loading it and then grepping anyway. Do not. Make the call explicitly:
| The question is… | Do this |
|---|---|
Where is X defined? What calls X? (you have the symbol name) |
graphify explain "X" — the cheapest, highest-precision call here. Measured 971 chars vs 1668 for the equivalent grep, and it reports call DIRECTION, which grep cannot. |
What breaks if I change X? |
graphify affected "X" (reverse traversal). |
How do A and B connect? |
graphify path "A" "B" — but empty output is NOT proof of no connection (dynamic imports, DI, string-keyed registries). |
| How is this codebase structured? Planning a cross-file refactor? | graphify query "<identifiers>" --budget 500 — see the budget warning below; never call query without --budget. |
| A literal string/comment/config value; a file whose path you already know; the exact current text of lines you are about to edit; a command's exit code; a field inside a JSON file | grep / read_text_file. The graph models code STRUCTURE — it does not model runtime values or literal text. Routing these through it returns confident noise. This is the right call, not a shortcut. |
Probe (graphify --version) failed, or no index exists and building is not appropriate here |
Fall back to grep, and say plainly the graph is unavailable. |
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.
- 1mo ago First seen · 142 lines · 77 tokens per session scan A fdb1cb74499b
graphify is a skill published in the GitHub repository aleksbuss/orchestra (2 stars, last pushed 13d ago), licensed MIT. It adds 77 tokens to every session and 2,349 once invoked, about $0.0003 per session on Opus 5.5. 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
triage
Triagem e causa raiz (RCA) de bugs. Use quando o usuário reportar "erro", "bug" ou "falha" para reproduzir, classificar e isolar o problema antes de qualquer correção.
perf-audit
Analisa gargalos de performance, complexidade algorítmica (Big-O) e vazamentos de memória. Use para "otimizar query", "reduzir alocação de memória" ou "auditar gargalos".
refactor
Reestruturação segura de código sem alterar o comportamento externo, para reduzir dívida técnica. Acionada pelo Architect ou Tech Lead.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
python-debugpy
Debug Python: pdb REPL + debugpy remote (DAP).
adk-debug
Diagnoses misbehaving ADK agents by inspecting sessions, events, tool calls, and the exact request that reached the model. Covers the adk run CLI and the adk web dev server with its session, trace, and debug HTTP endpoints. Use when an agent returns the wrong answer, ignores a tool or swallows a tool error, hangs…