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 joshrotenberg/hexpm-mcp --skill hexpmgit clone --depth 1 https://github.com/joshrotenberg/hexpm-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/skills/joshrotenberg/hexpm-mcp/hexpm)<a href="https://agentmods.dev/skills/joshrotenberg/hexpm-mcp/hexpm"><img src="https://agentmods.dev/badge/skills/joshrotenberg/hexpm-mcp/hexpm/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/joshrotenberg/hexpm-mcp/hexpm"><img src="https://agentmods.dev/badge/skills/joshrotenberg/hexpm-mcp/hexpm.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 | $0.00062 | $0.00522 |
| Opus 5.5 | $0.00025 | $0.00209 |
| Sonnet 5.5 | $0.00012 | $0.00104 |
| Haiku 4.5 | $0.00006 | $0.00052 |
Grade A, and why
hexpm 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 9d 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.
What it actually says
Hex dependencies with hexpm
The hexpm MCP server answers from hex.pm, HexDocs, OSV.dev and the Elixir Toolbox at call time. Versions, release dates, owners, advisories and APIs all change after training, so prefer the tools over memory for each of them.
Before adding or recommending a dependency
healthfor maintenance status, release cadence, downloads, owner count and retired versions.audit_mix_depswith that one dependency line, such as{:req, "~> 0.6"}, for the package's own advisories.healthdoes not query OSV, andauditchecks the package's dependencies rather than the package.auditwhen the package brings in dependencies the project does not already have.- When the result is weak (stale, one owner, open advisories), look wider with
alternativesortoolbox_search, thencomparetwo to five candidates.
Report what the tools returned, with versions and dates, not a verdict alone.
Upgrading
upgrade_checkwith the deps list frommix.exs. It flags major-version jumps.- For each upgrade,
releaseanddependenciesfor the target version, andversionsfor the releases in between. audit_mix_depson the updated list before reporting the upgrade done.
Writing code against a package
Read the API for the version in mix.lock instead of recalling it. Function
names, arities and options change between releases.
docswith that version for the module list.search_docsto find a module or function by name.doc_itemfor a module's full documentation.
Auditing a project
audit_mix_deps, then upgrade_check, each with the deps list from mix.exs.
audit_mix_deps counts every advisory filed against a package, whatever the
version. Before calling a dependency vulnerable, check that the version in
mix.lock falls inside the advisory's affected range.
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.
- 9d ago First seen · 50 lines · 62 tokens per session scan A 96db3574c497
hexpm is a skill published in the GitHub repository joshrotenberg/hexpm-mcp (2 stars, last pushed 7d ago), licensed MIT. It adds 62 tokens to every session and 522 once invoked, about $0.0002 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 2026-10-29.
hexpm compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| hexpmjoshrotenberg/hexpm-mcp | Skill | 2 | 7d ago | MIT | A | 62 | Claude Code |
| elixir-expertpersonamanagmentlayer/pcl | Skill | 44 | 13d ago | Apache-2.0 | A | 47 | Claude Code |
| elixir-reviewcamilooscargbaptista/cto-toolkit | Skill | 8 | 7mo ago | MIT | A | 25 | Claude Code |
| elixirericrisco/rsc-harness | Skill | 529 | 3d ago | MIT | A | 72 | Claude Code, Codex |
| tree-sitter-language-pack-extracting-code-structurexberg-io/tree-sitter-language-pack | Skill | 483 | yesterday | MIT | A | 52 | Claude Code |
| tree-sitter-language-pack-chunking-for-llmsxberg-io/tree-sitter-language-pack | Skill | 483 | yesterday | MIT | A | 52 | Claude Code |
| tree-sitter-language-pack-detecting-languagesxberg-io/tree-sitter-language-pack | Skill | 483 | yesterday | MIT | A | 55 | Claude Code |