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 protonium-labs/axiomcore-plugin --skill lintgit clone --depth 1 https://github.com/protonium-labs/axiomcore-pluginWrote 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/protonium-labs/axiomcore-plugin/lint)<a href="https://agentmods.dev/skills/protonium-labs/axiomcore-plugin/lint"><img src="https://agentmods.dev/badge/skills/protonium-labs/axiomcore-plugin/lint/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/protonium-labs/axiomcore-plugin/lint"><img src="https://agentmods.dev/badge/skills/protonium-labs/axiomcore-plugin/lint.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 | $0.00050 | $0.01258 |
| Opus 5 | $0.00025 | $0.00629 |
| Sonnet 5 | $0.00010 | $0.00252 |
| Haiku 4.5 | $0.00005 | $0.00126 |
Grade A, and why
lint 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task
Execute the requested action: $ARGUMENTS
Calling rule: /lint <action>.
If the requested action is ambiguous or contradictory, do NOT guess: flag it and ask for clarification before executing.
If action is empty
Describe this skill: its purpose and the actions table from "overview". Do not execute anything.
If action is "overview" (or unknown)
| Argument | Action |
|---|---|
all |
Markdown style check + vault health check + structure check |
markdown |
markdownlint across all .md files, auto-fix what is safe |
vault |
Full vault audit via the vault-gardener agent: freshness, links, orphans, index, structure mirror, frontmatter, contradictions |
structure |
Hierarchy: declared depth vs actual nesting, numbering, registry vs filesystem, naming convention |
overview |
This table |
If the action was unknown, say so first, then show the table.
If action is "markdown" (also part of "all")
- Run
npx --yes markdownlint-cli2 --fix "**/*.md"from the workspace root (config:.markdownlint.json, shipped with the workspace; if it is missing, create it with defaults and MD013 disabled). - Report anything not auto-fixed and fix those manually.
If action is "vault" (also part of "all")
- Delegate to the vault-gardener subagent. It owns the check list, the severity report, and the safe-fix rules.
- Relay its report to the user unchanged.
- For every "Needs your decision" item the user approves, apply the proposed fix, then re-run the affected check to confirm.
If action is "structure" (also part of "all")
Checks the hierarchy rules of root CLAUDE.md (Hierarchy Protocol, Output Hygiene & Naming, Task ID Protocol) against what is actually on disk. Registries are the source of truth: vault/projects/<project>/ids.md for projects, vault/workstations/<workstation>/areas.md for workstations.
- Depth declared. Every project and workstation must carry a
Levels: **2**orLevels: **3**line in its registry. Missing → report it; do not guess the depth from the folders. - Depth respected.
- 2-level: no sub-unit folders at all; every WP/WA sits directly under the project or workstation.
- 3-level: no WP/WA directly under the project or workstation; every one of them sits inside a numbered sub-unit.
A violation is never auto-fixed — moving a WP renumbers it and rewrites task IDs, which is a
/setupmigration. Report and stop there.
- Numbering. Every sub-unit, WP and WA folder carries a zero-padded numeric prefix. Numbers are unique within their parent — duplicates inside one sub-project are an error; the same number reused across two sub-projects is correct and expected at 3 levels.
- Naming convention. Every generated folder and file name is Latin-alphabet kebab-case (fixed-name and prefix exceptions per root CLAUDE.md). Report violations; renaming is a user decision.
- Registry vs filesystem. Every registry row has a matching folder and every folder a matching row, in both the working tree and the vault mirror. Each counter must exceed the highest number actually used — a counter that has fallen behind will hand out a duplicate on the next allocation.
- Mirror symmetry. For every WP and WA, the vault path matches the working-tree path exactly, acronym and number prefixes included.
- Task IDs match location. In each WP's
tasks.md, every task ID's prefix must equal that WP's own numbers (XX-WWat 2 levels,XX-SS-WWat 3) and its part count must match the project's declared depth. Report mismatches; renumbering is a migration, not a lint fix. - Work only in WP/WA. No
resources/oroutputs/folder anywhere except inside a WP or a work area. - Produced filenames. Every file you produced inside a WP or work area matches
<acr>-<address>-<NNN>-<name>.<ext>, with<acr>and<address>equal to the unit it sits in — a file carrying another unit's address is in the wrong folder. Setup files matchsu-<NNN>-<name>.md. Report: numbers duplicated within a unit (legitimate only for one document exported in several formats, same name part), aNext output/Next setupcounter at or below the highest number in use, leftover date or version strings in a filename, and markdown deliverables missing the header block or the## Changelogsection. Feature documents (<acr>-<address>-f###-<slug>.md) are exempt from the number, the header and the changelog. User uploads are never checked — only files you produced; when authorship is unclear, report nothing.
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 · 74 lines · 50 tokens per session scan A d0a948549d2f
lint is a skill published in the GitHub repository protonium-labs/axiomcore-plugin (3 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 1,258 once invoked, about $0.0003 per session on Opus 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
operating-cadence
Designs the rhythm an organization runs on — which reviews happen weekly, monthly and quarterly, what each one decides, who owns the numbers presented, and how a signal at the front line reaches the people who can act on it. Use this to set up a management operating system, fix a meeting calendar that produces no…
analyze
Deep-dive codebase analysis that explains how things actually work — business rules, architecture patterns, auth flows, data models, integrations, and performance hotspots. Use whenever the user asks "how does X work", "map the Y flow", "what are the business rules for Z", "trace the auth path", "explore the codebase…
debug
Systematically diagnose and resolve bugs through conversational investigation and root cause analysis.
testing
Writing effective tests and running them successfully. Covers layer-specific mocking rules, test design principles, debugging failures, and flaky test management. Use when writing tests, reviewing test quality, or debugging test failures.
performance-analysis
Measurement approaches, profiling patterns, bottleneck identification, and optimization guidance. Use when diagnosing performance issues, establishing baselines, identifying bottlenecks, or planning for scale. Always measure before optimizing.
markifact-overview
Reference — what Markifact is, what the MCP server exposes, and the discover→inspect→run pattern. Always loaded into the performance-marketer agent.