Home / cc-setup-tooling

Claude Code Setup Tooling

Fourteen skills that keep a Claude Code setup safe and tidy: review a third-party skill or plugin before installing it (skill-scan-gate, cc-plugin-lock and an offline inventory), lint SKILL.md front matter and fix its quoting, score descriptions against labelled prompts, estimate what a project loads on every turn, merge least-privilege permission rules and test them with claude-perm-sim, review the sandbox settings across scopes and flag what leaves it open, list the tool calls that would stop an unattended run on a permission prompt, show what each subagent can reach, add up the API usage in session transcripts and show which part moved, count shell commands that failed because of zsh, PowerShell or the Windows length limit, check what each compaction summary dropped, scaffold and test hooks with cc-hooks, find skills that collide or shadow each other, and find what breaks on other hosts and operating systems. Standard-library scripts, no network.

Plugin cc-setup-tooling, version 0.5.0, 14 skills, MIT licence. Source: claude-code-tooling-skills. Synced .

Install

In Claude Code, add the marketplace and install the plugin:

/plugin marketplace add basitalisandhu/claude-skills
/plugin install cc-setup-tooling@claude-skills

Or copy the skill files into ~/.claude/skills/ from a clone:

git clone https://github.com/basitalisandhu/claude-skills
cd claude-skills
python3 install.py --user --only cc-setup-tooling

Skills

SkillWhat it doesScripts
compaction-loss-checkCheck each compaction in a Claude Code session transcript: trigger, tokens before and after, time since the last user message, and whether the summary kept the standing rules the user gave, the files edited, branches and pull requests created and open to-dos.yes
context-budget-auditAudit what a Claude Code project puts in context on every turn (CLAUDE.md and its imports, AGENTS.md, rules files, skill descriptions, MCP tool schemas from saved tools/list output, hooks), estimate the tokens, rank the biggest items and flag duplicates and stale sections.yes
hook-authorWrite and test a Claude Code hook with cc-hooks: pick the event and the decision (block, warn, allow or modify) from a table, scaffold the hook, its fixtures, a pytest file and the settings block from a short spec, then prove it with cc-hooks test.yes
permissions-builderBuild a least-privilege .claude/settings.json permission set from what a project actually needs: generate MCP rules with claude-mcp-allow, merge them with your own allow, ask and deny lists, put deny first, diff against the current file, then test the result with claude-perm-sim against the commands the team runs.yes
sandbox-config-reviewReview a Claude Code project's sandbox settings across user, project, local and managed scopes, show what the operating system actually enforces around Bash commands, and flag what leaves it open: sandbox off, fail-open, the unsandboxed retry, broad excludedCommands, wildcard domains, readable credentials and loosening values a cloned repository ships.yes
session-usage-ledgerAdd up the API usage recorded in Claude Code session transcripts per day, session, model and project (calls, input, output, cache reads, 5-minute and 1-hour cache writes, cache-hit ratio, first-call size, subagent share) and compare two periods to show which part moved.yes
shell-slip-auditPair every Bash and PowerShell call in Claude Code session transcripts with its result and count the calls that failed because of the shell: zsh glob, equals and syntax errors, bash syntax sent to PowerShell, commands cut at the Windows length limit, including silent failures the call did not mark as errors.yes
skill-collision-checkFind installed skills that share a name, shadow each other or overlap so much in description that the wrong one gets picked, across personal, project, managed, plugin and copied folders, and say which one wins.yes
skill-description-linterLint the front matter of your SKILL.md files so each skill is listed and picked correctly: double-quoted description under 600 characters, a verb first, a quoted trigger phrase, Use when and Not for sentences, name equal to the folder, strict-YAML safety and a Limits section, with --fix for the quoting.yes
skill-portability-checkCheck a skill folder for what breaks on other hosts and operating systems: unquoted YAML colons, hard-coded home paths, CLAUDE_PLUGIN_ROOT with no fallback, GNU-only or BSD-only shell, Python open calls without encoding, POSIX-only calls, Windows path separators, CRLF and symlinks, with a fix hint per file.yes
skill-supply-chain-reviewReview a third-party skill or plugin before you install it: run skill-scan-gate on the folder, lock it with cc-plugin-lock, read every script for network and shell use with a bundled offline inventory, and write a one-page verdict.yes
skill-trigger-evalMeasure whether a skill's description triggers on the prompts it should and stays quiet on the ones it should not, using a labelled prompt set and transparent lexical scoring, and compare two description versions by precision and recall.yes
subagent-tool-scope-reviewReview every Claude Code subagent file for the tools it can actually reach: agents with no tools line inherit every built-in and MCP tool, lookalike fields such as allowed-tools are ignored, and reviewers that can write or run a shell.yes
unattended-run-preflightReplay the tool calls of a Claude Code run (a session log or a list) against the permission rules in your settings files and show which calls would stop a scheduled task, routine, CI job or claude -p run on a prompt or a deny rule, with the narrowest allow rule for each.yes

Plugin README

Skills that keep a Claude Code setup safe and tidy, each with a tested standard-library Python script. Install with /plugin marketplace add basitalisandhu/claude-code-tooling-skills and then /plugin install cc-setup-tooling@claude-code-tooling-skills. Skills appear as /cc-setup-tooling:<skill>, and Claude also invokes them on its own when a request matches a skill's description.

SkillScriptUse it to
skill-supply-chain-reviewskills/skill-supply-chain-review/scripts/skill_inventory.pyreview a third-party skill or plugin before installing it: skill-scan-gate, cc-plugin-lock scan and lock, and an offline inventory of scripts, imports, URLs, shell patterns, hooks, MCP servers and permissions, ending in a one-page verdict
skill-description-linterskills/skill-description-linter/scripts/description_lint.pylint SKILL.md front matter (double-quoted description under 600 characters, verb first, quoted trigger phrase, Use when, Not for, name equals folder, strict YAML, Limits section) and fix the quoting
skill-trigger-evalskills/skill-trigger-eval/scripts/trigger_eval.pyscore descriptions against a labelled prompt set with transparent lexical rules, report precision and recall, and compare two versions
context-budget-auditskills/context-budget-audit/scripts/context_budget.pyestimate what a project loads on every turn (memory files and imports, rules, skill listings, MCP tool schemas), rank it and flag duplicates and stale sections
permissions-builderskills/permissions-builder/scripts/perm_merge.pymerge allow, ask and deny rules from claude-mcp-allow and the team's own lists, order deny first, diff against the current settings, then test with claude-perm-sim
sandbox-config-reviewskills/sandbox-config-review/scripts/sandbox_review.pyreview the sandbox settings across user, project, local and managed scopes, show what is enforced and where each value comes from, flag what leaves the boundary open, and print curl checks to confirm enforcement
unattended-run-preflightskills/unattended-run-preflight/scripts/unattended_preflight.pyreplay a run's tool calls (session log or list) against the permission rules and list the calls that would stall a scheduled task, routine, CI job or claude -p run, with the narrowest allow rule for each
subagent-tool-scope-reviewskills/subagent-tool-scope-review/scripts/subagent_scope.pyshow which built-in tools and MCP servers each subagent can reach, and flag inherited access, ignored lookalike fields, readers that can write and bypass modes
session-usage-ledgerskills/session-usage-ledger/scripts/usage_ledger.pyadd up the API usage in session transcripts per day, session, model and project (calls, cache reads, 5-minute and 1-hour cache writes, cache-hit ratio, first-call size, subagent share) and compare two periods to see which part moved
shell-slip-auditskills/shell-slip-audit/scripts/shell_slips.pycount the Bash and PowerShell calls in session transcripts that failed because of the shell (zsh globs and syntax, PowerShell given bash, the Windows length limit), including silent ones, with fixes and a CLAUDE.md rule block
compaction-loss-checkskills/compaction-loss-check/scripts/compaction_check.pyfor each compaction, show the trigger, tokens and idle gap, check the summary for standing rules, edited files, branches, pull requests and open to-dos, and print what to re-state
hook-authorskills/hook-author/scripts/hook_scaffold.pyscaffold a cc-hooks hook from a short spec with fixtures, a pytest file and the settings block, then prove it with cc-hooks test
skill-collision-checkskills/skill-collision-check/scripts/skill_collisions.pyfind skills that share a name, shadow each other or overlap in description across managed, personal, project, plugin and copied folders, and say which wins
skill-portability-checkskills/skill-portability-check/scripts/portability_check.pyfind what breaks on other hosts and operating systems (YAML colons, home paths, plugin-root fallbacks, GNU or BSD shell, missing encoding, POSIX-only calls, backslash paths, CRLF, symlinks)

Requirements: Python 3.10 or newer on PATH as python3. The scripts read local files only: no network access, no third-party packages. The separate tools the skills drive (skill-scan-gate, cc-plugin-lock, cc-hooks, claude-perm-sim, claude-mcp-allow) are installed and run by the user; each skill says what to do when one is missing.