agentleFS
arbelharbelh/perdure/skills/health/SKILL.md

health

Skill0 starsChanged 30 days agoMITView on GitHub

Show perdure configuration health — the plugin-specific checks that

53 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use plan-tune, ios-clean and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ShreyRaviSkill

    plan-tune

    Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required.

    Also in 9 projects that use this

  • Saqlain-811Skill

    ios-clean

    Remove the DebugBridge SPM package and all #if DEBUG wiring from an iOS app. (gstack)

    Also in 5 projects that use this

  • Saqlain-811Skill

    ios-fix

    Autonomous iOS bug fixer. (gstack)

    Also in 5 projects that use this

  • seek888Skill

    pair-agent

    Pair a remote AI agent with your browser. One command generates a setup key and prints instructions the other agent can follow to connect. Works with OpenClaw, Hermes, Codex, Cursor, or any agent that can make HTTP requests. The remote agent gets its own tab with scoped access (read+write by default, admin on request). Use when asked to "pair agent", "connect agent", "share browser", "remote browser", "let another agent use my browser", or "give browser access". (gstack) Voice triggers (speech-to-text aliases): "pair agent", "connect agent", "share my browser", "remote browser access".

    Also in 11 projects that use this

  • ShreyRaviSkill

    landing-report

    Read-only queue dashboard for workspace-aware ship. Shows which VERSION slots are currently claimed by open PRs, which sibling Conductor workspaces have WIP work likely to ship soon, and what slot /ship would pick next. No mutations — just a snapshot. Use when asked to "landing report", "what's in the queue", "show me open PRs", or "which version do I claim next". (gstack)

    Also in 8 projects that use this

  • timurgaleevSkill

    document-generate

    Generate missing docs for a feature, module or project as tutorials, how-tos, reference and explanation.

    Also in 7 projects that use this

  • +1 more from Saqlain-811
  • +2 more from ShreyRavi

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 38 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    READ-ONLY report on the plugin-specific configuration layer.

    line 15
  2. Very specific:

    Read the installed plugin's version (the version in ~/.claude/plugins//perdure*/.claude-plugin/plugin.json, or via claude plugin list) and …

    line 23
  3. Distinctive:

    If the installed version is older, lead the report with the nudge:

    line 28
  4. Very specific:

    When Codex's folder exists, Codex gets the records rule from either of two places: ~/.codex/AGENTS.md carrying the same marker block (Codex …

    line 49
  5. Distinctive:

    If in a project directory:

    line 89
  6. Very specific:

    NEVER modify files; reading files, printenv CODEX_HOME and claude plugin list is the entire surface

    line 106

“Common” is measured: kept by at least 6 in 10 of the 38 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

perdure health check

READ-ONLY report on the plugin-specific configuration layer. First-party tools already cover generic health (/doctor, /plugin list, /agents) — this skill checks only what THEY can't see:

Checks

1. Version lag (installed vs available)

Read the installed plugin's version (the version in ~/.claude/plugins/**/perdure*/.claude-plugin/plugin.json, or via claude plugin list) and compare against the newest version available to this machine — the local marketplace clone's copy (~/.claude/plugins/marketplaces/*/).

If the installed version is older, lead the report with the nudge:

⚠ Installed perdure v<X> lags available v<Y>.
  Update: claude plugin update perdure@<marketplace>   (then restart the session)

Silent version lag is a known failure mode (an earlier install once ran 3 months stale) — this check exists so it can't happen quietly.

2. Personal configuration (~/.claude/, Codex's folder)

Codex's folder is $CODEX_HOME when that variable is set, else ~/.codex (printenv CODEX_HOME says which). Below, ~/.codex/AGENTS.md means AGENTS.md in that folder.

  • ~/.claude/CLAUDE.md contains the <!-- perdure-records:begin --> … <!-- perdure-records:end --> marker block. Without it, look for a records rule of the user's own (a line that names /perdure:start or perdure/workflows/): report it as found outside the markers, with its line numbers, not as missing. Both present: say so, since two rules can disagree.
  • When Codex's folder exists, Codex gets the records rule from either of two places: ~/.codex/AGENTS.md carrying the same marker block (Codex reads that file in every project), or perdure's session-start hook (4.4.0 and later) once the user has trusted it: a [hooks.state] entry naming codex-session-start.json in Codex's config.toml that has a trusted_hash and does not say enabled = false. Either → ✓, naming which; the hook stays silent where the file carries the rule. For the hook, add that the hash is not compared with the current hook, so a hook changed since it was trusted still reads ✓ here while Codex asks to review it again. Neither → ✗ with the fixes: trust the hook in Codex's /hooks, or /perdure:install. A records rule of the user's own there is reported as in ~/.claude/CLAUDE.md. When a non-empty AGENTS.override.md sits in the same folder, say that Codex reads it instead of AGENTS.md, so the block is not read while the override exists. No Codex folder: skip this line.
  • ~/.claude/perdure-preferences.json exists. Show:

    • the active values. A key the file lacks takes the value the code reads while it is missing; say which keys, and that value: auto-promote-decisions heuristic, auto-archive-before-compact off (the hook archives only on an explicit true), rot-lint-at-close on, handoff-regen-at-close on, commit-archives off
    • the ones that differ from today's defaults, each with the default beside it. Defaults, as /perdure:install writes them: auto-promote-decisions: heuristic, auto-archive-before-compact: true, rot-lint-at-close: true, handoff-regen-at-close: true, commit-archives: false
    • keys the current schema ignores, listed by name (workflow-files, verbose-status, parallel-exploration, orchestration-mode, suggest-agent-skills, auto-reindex, archive-location, headless-budget-usd, headless-result-inline-threshold, agent-permissions-approved, or any other key the schema does not name)
    • its schema-version, against today's 4

    A value that differs is a choice, not a fault: mark it ○. To move to the defaults, /perdure:install offers an update and keeps the file on enter. With no preferences file, every key takes its while-missing value: say that nothing is archived before compaction.

3. Current project records

If in a project directory:

  • perdure/workflows/ — count files by status (scan frontmatter; no index file exists)
  • perdure/decisions/ — count ADRs
  • perdure/archives/ — count archives; note whether .gitignore covers them and whether that matches the commit-archives preference

4. Report

Plain markdown, short sections, symbols ✓ configured / ✗ missing (with the fix command) / ○ optional. End with either the prioritized fix list (what to run, what it changes) or:

🎉 Fully configured — nothing to fix.

Safety rules

  • NEVER modify files; reading files, printenv CODEX_HOME and claude plugin list is the entire surface
  • Missing/unreadable file → ✗, not an error
  • Show absolute paths so the user can investigate
  • Outside a project directory, skip the project section

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.