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 alex-feel/claude-code-artifacts-public --skill serena-tool-selectiongit clone --depth 1 https://github.com/alex-feel/claude-code-artifacts-publicWrote 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/alex-feel/claude-code-artifacts-public/serena-tool-selection)<a href="https://agentmods.dev/skills/alex-feel/claude-code-artifacts-public/serena-tool-selection"><img src="https://agentmods.dev/badge/skills/alex-feel/claude-code-artifacts-public/serena-tool-selection/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/alex-feel/claude-code-artifacts-public/serena-tool-selection"><img src="https://agentmods.dev/badge/skills/alex-feel/claude-code-artifacts-public/serena-tool-selection.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 | $0.00193 | $0.04921 |
| Opus 5.5 | $0.00077 | $0.01968 |
| Sonnet 5.5 | $0.00039 | $0.00984 |
| Haiku 4.5 | $0.00019 | $0.00492 |
Grade A, and why
serena-tool-selection 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 25d 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 — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Serena Tool Usage
When Serena tools are available, use them for code navigation and symbol-level editing inside the project Serena serves, in preference to Search, Grep, and Read: Serena resolves definitions and usages semantically, following aliases and renamed imports that a text search misses, so a text search for a symbol can return an incomplete set even when it looks faster or simpler. The built-in tools remain the right choice for the cases listed under Step 3, and as the fallback when Serena is unavailable or cannot resolve a symbol.
Symbol Searches to Route to Serena
Each row below names a text-search habit and the Serena tool that answers the same question completely, with the canonical call. Every path in a Serena call is relative to the project root, as described under "Name Paths and Relative Paths" below.
| Instead of | Use Instead | Canonical Call |
|---|---|---|
Search(pattern: "def process_data", path: "/project/src") or Grep(pattern: "def process_data") -- any Search/Grep pattern that is a function/class/method name |
find_symbol (definitions) or find_referencing_symbols (usages) |
find_symbol(name_path_pattern="process_data", include_body=True) |
Grep(pattern: "validate_input\\(...") or Search(pattern: "validate_input(", output_mode: "content") -- any search for a symbol's usages/calls |
find_referencing_symbols + Grep cross-validation when completeness matters |
find_referencing_symbols(name_path="validate_input", relative_path="src/validation.py") |
Read entire file to understand structure |
get_symbols_overview |
get_symbols_overview(relative_path="src/module.py") |
Multiple Edit calls to rename a symbol |
rename_symbol |
rename_symbol(name_path="old_func", relative_path="src/module.py", new_name="new_func") |
Finding function boundaries then Edit |
replace_symbol_body |
replace_symbol_body(name_path="fn", relative_path="src/module.py", body="def fn():\n return 42") |
Finding method end line then Edit |
insert_after_symbol |
insert_after_symbol(name_path="existing", relative_path="src/module.py", body="def new():\n pass") |
find_referencing_symbols then manually checking results and deleting via Edit -- manual ref-check + Edit to delete |
safe_delete_symbol |
safe_delete_symbol(name_path_pattern="old_handler", relative_path="src/handlers.py") |
Separate Edit calls file by file across the project -- repeating an identical Edit across many files for one textual change |
replace_in_files (dry run first) |
replace_in_files(needle="old_key", repl="new_key", mode="literal", dry_run=True), review the diff, then apply with expected_count |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 25d ago Changed · -2 lines · -3 tokens per session 2985480f57de
- 29d ago Changed · +8 lines · +16 tokens per session 559efae3a12d
- 1mo ago Changed · -2 lines · +117 tokens per session eef0da0c192e
- 1mo ago Changed · +21 lines · +9 tokens per session fa6e554090b4
- 2mo ago First seen · 162 lines · 54 tokens per session scan A 129f0a27d4fb
serena-tool-selection is a skill published in the GitHub repository alex-feel/claude-code-artifacts-public (6 stars, last pushed 2d ago), licensed MIT. It adds 193 tokens to every session and 4,921 once invoked, about $0.0008 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
prod-debug
Production debugging skill. Pre-loads DB schema, container registry, and prod environment facts from .claude/prod-debug/ in the current project root. Use when the user invokes /prod-debug or /prod-debug bootstrap.
sysops
Invoke the sysops subagent for VPS maintenance. Use when the user runs /sysops with an optional argument like "status", "update Pulse", or "update my servers".
ralph-config
Generate ralph.yml and PROMPT.md configuration files for Ralph Orchestrator workflows from plans, tickets, or ideas. Use when setting up Ralph Orchestrator, creating hat-based workflows, or adapting Linear/Jira/markdown plans into ralph.yml + PROMPT.md files.
breaking-down-work
Breaks down epics into features and tasks optimized for AI agent execution. Includes structure preparation, parallel codebase analysis, feature sizing, task generation, and dependency mapping. Use when an epic is ready for sprint planning or when a large work item needs decomposition into actionable pieces.
building-skills
Build production-ready Claude Code skills through a structured Research-Draft-Refine process. Use when creating new skills, converting slash commands to skills, or generating AgentSkills.io-compatible skills from specifications.
refining-work-items
Refines work items (epics, features, tasks, bugs) into AI-ready specifications. Use when converting rough ideas into structured items, preparing epics for breakdown, enhancing existing items with acceptance criteria, or creating new items from scratch. Supports --type (epic|epic-lite|feature|task|bug)…