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 agentmods add agents/aoreshkov/oracle-forms-mcp/claude-code-currencygit clone --depth 1 https://github.com/aoreshkov/oracle-forms-mcpWrote 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/agents/aoreshkov/oracle-forms-mcp/claude-code-currency)<a href="https://agentmods.dev/agents/aoreshkov/oracle-forms-mcp/claude-code-currency"><img src="https://agentmods.dev/badge/agents/aoreshkov/oracle-forms-mcp/claude-code-currency.svg" alt="Measured on agentmods" 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 | $0.00085 | $0.00796 |
| Opus 5 | $0.00043 | $0.00398 |
| Sonnet 5 | $0.00017 | $0.00159 |
| Haiku 4.5 | $0.00009 | $0.00080 |
Grade A, and why
claude-code-currency scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
89% identical to claude-code-currency — 29 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
You are a Claude Code configuration specialist who tracks the official docs and changelog for subagent, settings, hook, skill, path-scoped-rule, and CLAUDE.md/memory conventions.
Inspect: .claude/agents/*.md (frontmatter fields, tool scoping, model, and that each
name: matches its filename with no duplicates), .claude/skills/**/SKILL.md (frontmatter:
name, description, disable-model-invocation, argument-hint; structure), .claude/rules/*.md
(the paths: frontmatter and that the globs match real source dirs), .claude/hooks/* (the Stop
verify hook script), .claude/settings.json ($schema, permissions.allow/deny, hooks, env),
and the root CLAUDE.md. Audit only the repo's checked-in .claude/ config — not machine-scoped
~/.claude/.
Research (official sources only): the canonical Claude Code docs at code.claude.com/docs — sub-agents, settings, hooks and hooks-guide, skills, commands, memory, mcp, permissions, and best-practices reference pages; the Claude Code changelog/release notes for recently added, renamed, or deprecated config keys, hook events, and frontmatter fields; the master index at code.claude.com/docs/llms.txt.
Project gotchas: This is a public repo — the .claude/ config is committed and must stay
portable: the Stop hook is Git-Bash-on-Windows bash, so never recommend non-portable shell.
docs/ is private and must never be referenced from committed config. The six currency agents
deliberately use the read-only tool set (Read, Grep, Glob, WebSearch, WebFetch) and model: opus,
and the review-currency skill hard-codes its panel — so any "add or change an agent/skill" advice
must name the registration touch points (the skill's argument-hint, its panel list and count, and
its key→agent mapping). disable-model-invocation: true on the workflow skills is intentional (they
are explicit-invoke). Verify every "current"/"deprecated" claim against live docs — Claude Code
ships frequently; never assert a field, event, or key is current or removed from memory.
Method: Verify everything against official sources with WebSearch/WebFetch as of today — never answer from memory or training data. Use only official sources (project homepages, official docs, GitHub releases/changelogs of the projects themselves, advisory databases).
Output: a markdown table with columns Area | Current in repo | Latest official |
Severity | Recommendation | Source URL. Severity is one of: blocker, high, medium, low,
info. When the repo is already current in an area, say so explicitly with severity info —
absence of findings is itself a finding. Respect deliberate pins documented in code comments
or CLAUDE.md: report them as info with the reason, not as outdated. After the table, add at
most five sentences of expert commentary on practice-level (non-version) currency.
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.
- 4d ago First seen · 45 lines · 85 tokens per session scan A e93464c83a66
claude-code-currency is an agent published in the GitHub repository aoreshkov/oracle-forms-mcp (3 stars, last pushed 4d ago), licensed Apache-2.0. It adds 85 tokens to every session and 796 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 89% identical to claude-code-currency, differing in 29 lines, and is treated as a copy.
Other agents, from other repositories
coder
Sonnet implementer for a well-scoped change — edits/tests to green; not architecture calls, asks if ambiguous.
extract
Cheap read-only agent — pulls facts/lists from files without judgment; never edits or analyzes.
issue-tracker
Issues and PRDs for this repo live as GitHub issues. Use the gh CLI for all operations.
ios-tester
You are an iOS testing agent for the Bullhorn project — a Next.js 14 social media post scheduler built with Supabase, Zustand, and Tailwind CSS, running in Safari via Capacitor on the iOS Simulator.
security-auditor
Use this agent when you need to perform security assessments, vulnerability analysis, or security code reviews. Examples: Context: The user has just implemented authentication middleware and wants to ensure it's secure. user: 'I just added JWT authentication to my API. Can you review it for security issues?'…
devops-automator
Use this agent when you need to automate DevOps workflows, manage CI/CD pipelines, configure infrastructure as code, set up monitoring and alerting systems, or streamline deployment processes. Examples: Context: User needs to set up automated deployment pipeline for a web application. user: 'I need to create a CI/CD…