status

status is a command for coding agents from biznuts/never-stale. It costs 64 tokens per session (2,315 once invoked), scanned A, original, MIT.

A read-only health check for the never-stale tool in the current project. It reports which project marker applies, whether the tool is enabled, what languages are recorded, and whether its change-detection gate would run.

In plain words
What is it for?
Checking project configuration, comparing team and local settings, inspecting the CLAUDE.md marker state, and diagnosing why the gate would or would not fire.
Why use it?
It answers whether never-stale is active and what it would do without changing project files.

Command

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the never-stale plugin — 8 commands, 2 hooks shipped together

Install

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.

agentmods
npx agentmods add commands/biznuts/never-stale/status
Clone the repo
git clone --depth 1 https://github.com/biznuts/never-stale

Or install never-stale, the plugin that ships this one along with the rest of its 8 commands, 2 hooks.

Wrote 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.

agentmods badge for status

README.md
[![agentmods](https://agentmods.dev/badge/commands/biznuts/never-stale/status.svg)](https://agentmods.dev/commands/biznuts/never-stale/status)
Your own site
<a href="https://agentmods.dev/commands/biznuts/never-stale/status"><img src="https://agentmods.dev/badge/commands/biznuts/never-stale/status.svg" alt="Measured on agentmods" height="20"></a>
Per session 64 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,315 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00064 $0.02315
Opus 5 $0.00032 $0.01157
Sonnet 5 $0.00013 $0.00463
Haiku 4.5 $0.00006 $0.00231

Measured 3d ago against content hash feca800dc066, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

status 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 3d 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.

never-stale/commands/status.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

never-stale — status of this project

A read-only health check for the current project. It writes nothing — it inspects and reports. Use it to answer "is never-stale on here, and what would the gate do?"

Project root = the path in $ARGUMENTS if given, else the current working directory. Call it <ROOT>.

Step 1 — Resolve the governing marker (the gate's own walk)

Walk up from <ROOT> to the nearest ancestor that carries a marker, bounded by the git repo root:

  1. At each directory D, check D/.claude/never-stale.local.json then D/.claude/never-stale.json.
  2. If either exists, that directory governs — stop. (local overrides team at the same directory — but a local file that fails to PARSE is ignored in favour of the team one, so a typo'd local marker never silently disables a team opt-in.)
  3. Otherwise, if D/.git exists, stop (repo root — no governing marker).
  4. Else go to the parent and repeat.

Record the governing directory <GOV>, which marker is effective (local vs team), and whether the launch dir is the marker's own directory or a descendant (the upward walk).

Step 2 — Inspect the artifacts

Read the installed plugin version from ${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json (version); call it <PV>. Use it for the drift checks below. If you cannot resolve ${CLAUDE_PLUGIN_ROOT}, skip the version-drift line rather than guessing.

  • Marker — read the effective marker. Report: scope (team/local), enabled (true / false / invalid if the JSON is corrupt — the gate treats corrupt as disabled, except that a corrupt local marker falls back to the team marker beside it), recorded spoken / written (with spokenCode / writtenCode if present), per-event flags (events.compact, events.edit; absent = on), version, createdAt. If both a team and a local marker exist at <GOV>, note that the local one wins when it parses and show both.
    • Version drift — if version !== <PV>, note it: "marker written by <version>, plugin is <PV>." Make clear this is cosmetic — the gate ignores the version — and that /never-stale:update reconciles it.
    • Language codes — if spokenCode / writtenCode are missing, or the display string is non-canonical (not one of English / Traditional Chinese / Traditional Chinese (Hong Kong) / Simplified Chinese, nor a deliberate "Other"), note it and point at /never-stale:update to normalize. Also informational only.
    • Drift pairs (syncPairs) — if the marker declares syncPairs, show a resolution preview for each pair so a misconfiguration is visible rather than silent (the gate ignores a pair it cannot resolve). For each { source, snapshot, mode }:
      • resolve both paths relative to <GOV>; flag any that is unsafe (absolute / drive-qualified / UNC / parent-escaping .., or resolving through a symlink/junction to outside the repo) or missing on disk — the gate silently skips those;
      • for mode: "mtime" (the default), read both files' mtimes and state the verdict: drift (source newer than snapshot → "snapshot may be behind, verify") or clean (snapshot at least as new);
      • for mode: "hash" (v0.10.0+), read the snapshot's synced-to marker (<!-- never-stale:synced-to <hex> -->) and the source's current normalized hash, then state the verdict: drift (the marker is not a prefix of the current source hash → "snapshot recorded <declared>, source is now <current>; reconcile and update the marker"), clean (they match), or unknown (no synced-to marker in the snapshot, or the source is missing / larger than the 512 KB cap — the gate skips these silently, so surfacing them here is the whole point);
      • for mode of declared / version, note it is deferred — not planned (a reserved value the gate treats as a silent no-op).

Read the full file on GitHub · 128 lines

Changes

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.

  1. 3d ago First seen · 128 lines · 0 tokens per session scan A feca800dc066

Subscribe to this mod's changes

status is a command published in the GitHub repository biznuts/never-stale (2 stars, last pushed 1mo ago), licensed MIT. It adds 64 tokens to every session and 2,315 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.