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 urmzd/dotfiles --skill align-readmesgit clone --depth 1 https://github.com/urmzd/dotfilesWrote 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/urmzd/dotfiles/align-readmes)<a href="https://agentmods.dev/skills/urmzd/dotfiles/align-readmes"><img src="https://agentmods.dev/badge/skills/urmzd/dotfiles/align-readmes/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/urmzd/dotfiles/align-readmes"><img src="https://agentmods.dev/badge/skills/urmzd/dotfiles/align-readmes.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 | $0.00113 | $0.01312 |
| Opus 5.5 | $0.00045 | $0.00525 |
| Sonnet 5.5 | $0.00023 | $0.00262 |
| Haiku 4.5 | $0.00011 | $0.00131 |
Grade A, and why
align-readmes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Align READMEs
One fleet, one voice. This skill owns the cross-repo loop. It does not own README rules: those live in write-readme (structure) and clean-docs (prose). Never restate them here.
Usage: /align-readmes [root] (default root: ~/github)
1. Triage before touching anything
Run the bundled audit. It is read-only.
bash "${CLAUDE_SKILL_DIR:-$HOME/.agents/skills/align-readmes}/scripts/audit-readmes.sh" ~/github
Act on the status and identity columns first:
| Value | Meaning | Action |
|---|---|---|
ARCHIVED |
Read-only on GitHub | Skip. Never edit. List it in the report so the user can delete the local checkout. |
MISSING |
Remote repo deleted or inaccessible | Skip. Ask the user. |
no-remote |
Local-only directory | Skip unless the user names it. |
dir-is:<name> |
Repo was renamed, folder was not | Use <name> in every README edit. Do not rename the folder without asking. |
stale-remote:<o/r> |
Remote URL predates a rename | Report git remote set-url origin fix. Do not run it unasked. |
Also verify against GitHub, not memory: renames happen in waves and recall goes stale.
gh repo list urmzd --limit 300 --json name,isArchived --jq '.[] | select(.isArchived) | .name'
2. Score the active set
Remaining columns map to write-readme rules:
| Column | Rule owner | Pass |
|---|---|---|
hdr |
write-readme: Centered Header | 3 |
badges |
write-readme: Badges | 1 to 4 |
em_dash |
clean-docs | 0 |
tree |
write-readme: No Project Structure | 0 |
toc_needed |
write-readme: ToC (300+ lines) | ok or - |
quick_start |
write-readme: Quick Start | ok |
skill_2nd |
write-readme: Agent Skill position | ok, or no when the repo has no skills/ |
license_last |
write-readme: Section Order | ok |
The script is a fast filter, not the verdict. Read each failing README before editing.
Exempt: the urmzd profile repo is a GitHub profile page, not a project README. Skip header, Quick Start, and section-order rules for it.
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.
- 4d ago First seen · 110 lines · 113 tokens per session scan A 5eae204b15b0
align-readmes is a skill published in the GitHub repository urmzd/dotfiles (5 stars, last pushed 5d ago), licensed Apache-2.0. It adds 113 tokens to every session and 1,312 once invoked, about $0.0005 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 2026-10-21.
Other skills, from other repositories
mcp-notion-usage-guide
A usage guide for connecting AI agents to Notion databases through MCP. Notion is a workspace for documents and databases, while a database view is a saved way of displaying database records.
orchestrate-rex
Oversee a batch of related work items (tickets, PRs and what blocks them) through to done from one session. Keeps a status table, sets the order, watches slow external runs, and starts Rex sessions for workers when code needs changing. Use when the user hands over a batch of work to coordinate.
write-ticket-description
Voice guide for writing issue/ticket/task/epic descriptions. Use this skill for ALL issue creation and for every rewrite of an existing issue's body — GitHub issues, Jira tasks, Monday tasks, Linear tasks, and epics. Contains section structure, voice rules, and anti-patterns. Never write ticket descriptions without…
implement-unattended
Take a ticket whose approach and design are already approved through implementation, review and evidence to a draft PR, with nobody watching. Escalates decisions on the ticket and stops rather than choosing.
plan-unattended
Plan a ticket with nobody watching. Three independent discuss runs are reconciled into one approach and three design runs into one design, both are posted on the ticket, and implement-unattended is started in a new tmux window. Escalates on the ticket rather than taking a one-way decision.
wrap-up
End a session by confirming every action it mentioned was dealt with, then finding what in the codebase and the backlog would mislead the next agent. Reports both and fixes only what the user approves.