suboss87 · suboss87 · Aug 27, 2026 · Aug 27, 2026
@@ -5,13 +5,13 @@
     "url": "https://github.com/suboss87"
   },
   "metadata": {
-    "description": "Second brain for Forward Deployed Engineers - one @fde skill, phase methods, and local .fde/ engagement memory."
+    "description": "Engagement memory for AI coding agents - one @fde skill, phase methods, and local .fde/ client memory."
   },
   "plugins": [
     {
       "name": "fdeops",
       "source": "./",
-      "description": "Second brain for Forward Deployed Engineers: one @fde entry point that routes by situation (land, discover, build, rescue, ship, close), does the phase work, and keeps per-customer engagement memory in local .fde/ files. Healthcare, fintech, and government overlays included."
+      "description": "Engagement memory for AI coding agents. One entry point that routes by situation (land, discover, build, rescue, ship, close), does the phase work, and keeps per-client memory - sponsor, promise, decision, acceptance, dated - in local .fde/ files. For Forward Deployed Engineers running several clients at once; healthcare, fintech, and government overlays included."
     }
   ]
 }
@@ -1,7 +1,7 @@
 {
   "name": "fdeops",
-  "description": "Second brain for Forward Deployed Engineers. One @fde skill - enter at day one, mid-project, or mid-fire; it routes and does the phase work; engagement memory lands in .fde/ as you confirm judgment.",
-  "version": "3.11.1",
+  "description": "Engagement memory for AI coding agents. One @fde skill routes the client work - land, discover, plan, build, ship, close - and the record of it (sponsor, promise, decision, acceptance, dated) lands in local .fde/ files as you confirm judgment. For Forward Deployed Engineers running several clients at once.",
+  "version": "3.12.0",
   "category": "productivity",
   "tags": [
     "community-managed"
@@ -14,6 +14,8 @@
   "repository": "https://github.com/suboss87/fdeops",
   "license": "MIT",
   "keywords": [
+    "engagement-memory",
+    "agent-memory",
     "fde",
     "forward-deployed",
     "legacy-code",

@@ -1,5 +1,18 @@
 # Changelog
 
+## 3.12.0 — 2026-08-27
+
+Vocabulary: standard words on the outside, so nothing has to be learned before it works.
+
+### Changed
+- **The skill fires on client work, not on a password.** The frontmatter `description` - the only text a host reads before loading a skill - triggered on `@fde` plus a list of our own nouns ("hygiene", "receipts"), so an FDE who simply talked about their client got no memory. It now names the intents in the standard `Use when …` form: mentioning a client or sponsor, debriefing a meeting, asking what was agreed, prepping a readout, opening a session in a client workspace. `@fde` remains, as one trigger among several rather than the gate.
+- **`fde tidy` is the verb; `fde garden` still works.** "Garden" was a metaphor only we used. Nothing is removed - the old name routes to the same code, and the alias is covered by a test.
+- **`capture` and `preserve` left the human help.** Only hooks call them; listing them invited an FDE to run a snapshot by hand.
+- **"Engagement memory for AI coding agents"** replaces "second brain" in the npm, plugin and marketplace descriptions - the phrase someone would actually search, and an accurate one.
+
+### Added
+- **One glossary.** The six words that carry the method (fieldbook, brief vs reality, terrain, trust signal, receipts, vault) are defined once in the README instead of being met scattered and guessed. A gate keeps it in place, and a second gate keeps the description triggering on intent.
+
 ## 3.11.1 — 2026-08-27
 
 Adoption: one skill, and the CLI is never missing.

@@ -7,9 +7,9 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
 [![Node](https://img.shields.io/badge/node-%3E%3D18-brightgreen)](https://nodejs.org)
 
-Skill packs teach your AI coding agent how to *build*. None of them remember who the client is, what you promised, or who agreed it was delivered. FDEOps adds that layer: a private fieldbook per engagement (`.fde/`), a field methodology from land to close, and one `@fde` skill that routes it.
+**Engagement memory for AI coding agents.** Skill packs teach your agent how to *build*. None of them remember who the client is, what you promised, or who agreed it was delivered - close the window and it is gone. FDEOps adds that missing layer: a private record per client (`.fde/`), a field methodology from land to close, and one skill that routes it.
 
-Built for Forward Deployed Engineers and anyone embedded in client work - consultants, agency developers, solutions architects, fractional CTOs. Feels like a second brain; behaves like a defensible record: dated, sourced, on your laptop.
+Built for Forward Deployed Engineers and anyone embedded in client work - consultants, agency developers, solutions architects, fractional CTOs. Not notes: a defensible record - dated, sourced, on your laptop.
 
 ```
   land      discover      plan      build      ship      close
@@ -58,9 +58,11 @@ npx fdeops resume                 # where we are
 **3. Work** in plain language:
 
 ```text
-@fde New client. Payments platform. They want it live before the Q3 audit.
+New client, Garvey. Payments platform. They want it live before the Q3 audit.
 ```
 
+No prefix needed - naming a client is enough. Say `@fde` when you want it explicitly.
+
 **It is working if** `npx fdeops resume` prints this client's phase, trust signal, and next action - and prints the same thing tomorrow, from a new session, with no explaining. Full workflow: [docs/USAGE.md](docs/USAGE.md).
 
 <details>
@@ -79,16 +81,16 @@ npx fdeops resume                 # where we are
 
 ## The week
 
-`@fde` plus English. No cheat sheet.
+**Say what happened, in your words.** There is nothing to memorise - no flags, no cheat sheet. The skill picks it up when you mention a client, debrief a meeting, or ask what was agreed; `@fde` is only the explicit way to summon it.
 
 | When | What you say | What you get |
 |------|--------------|--------------|
 | **Start of week** | `@fde` - or just open Claude Code | Fieldbook on disk either way. **Claude Code** injects trust, phase, next before you type. **Cursor / Codex / others:** say `@fde` or `resume` - nothing auto-loads. |
-| **After a meeting** | `@fde` debrief these notes *(paste or attach)* | Proposed updates. You review, then confirm. |
-| **Optional: pull** | `@fde` connect Granola *(once)* · `@fde` pull today's Acme transcript | You add that source MCP. We **pull** on request - no push, no sync. [mcp/recipes/](mcp/recipes/) |
-| **Before a stakeholder meeting** | `@fde` prep me for tomorrow with the sponsor | Brief from what you already logged. |
-| **Scope dispute** | `@fde` when did we agree to drop that? | Dated answers, or a clear gap. |
-| **End of week** | `@fde` draft the sponsor update from the record | Status grounded in what happened. |
+| **After a meeting** | here are my notes from the Acme call *(paste or attach)* | Proposed updates. You review, then confirm. |
+| **Optional: pull** | connect Granola *(once)* · pull today's Acme transcript | You add that source MCP. We **pull** on request - no push, no sync. [mcp/recipes/](mcp/recipes/) |
+| **Before a stakeholder meeting** | prep me for tomorrow with the sponsor | Brief from what you already logged. |
+| **Scope dispute** | when did we agree to drop that? | Dated answers, or a clear gap. |
+| **End of week** | draft the sponsor update from the record | Status grounded in what happened. |
 
 Same folder every time: `~/fde-engagements/<client>/.fde/`.
 
@@ -149,6 +151,8 @@ One folder per client. Plain markdown, so you can grep it, diff it, copy it into
 
 A day-one fieldbook ships **empty** - headings and allowed values, no invented rows - so anything you read in it is something that actually happened. Schema: [docs/schema.md](docs/schema.md).
 
+**Words used here, once:** *engagement* - one client's body of work, one folder. *Fieldbook* - that folder (`.fde/`), the record itself. *Brief vs reality* - what they said the problem was, and what it turned out to be. *Terrain* - their systems and org as you actually found them. *Trust signal* - green / amber / red on one relationship. *Receipts* - the dated line proving something was agreed. *Vault* - the Obsidian copy `fde vault` generates to read it all in one window.
+
 ---
 
 ## The field methods

@@ -286,6 +286,33 @@ if (!read('skills/fde/SKILL.md').includes('npx --yes fdeops')) {
   fail('skills/fde/SKILL.md must fall back to npx --yes fdeops when the CLI is not installed')
 } else ok('SKILL.md npx CLI fallback')
 
+// The frontmatter description is the only text every host reads before deciding
+// to load the skill. If it triggers on "@fde" alone, an FDE who just talks about
+// their client gets no memory - so it must carry natural-intent triggers in the
+// "Use when …" convention, and @fde must be one of several, never the gate.
+{
+  const fm = /^---\n([\s\S]*?)\n---/.exec(read('skills/fde/SKILL.md'))
+  const desc = fm ? (/^description:[^\S\n]*(.*)$/m.exec(fm[1]) || [])[1] || '' : ''
+  const triggers = desc.match(/Use when/g) || []
+  if (triggers.length < 4) {
+    fail(`SKILL.md description needs several "Use when …" triggers so it fires on intent (found ${triggers.length})`)
+  } else if (/Use when the human says @fde or/.test(desc)) {
+    fail('SKILL.md description must not gate on @fde - name the client-work intents first')
+  } else ok('SKILL.md description triggers on intent')
+}
+
+// Six invented words (fieldbook, terrain, reality, trust signal, receipts,
+// vault) carry the method. They were never defined in one place, so a stranger
+// met them scattered through the docs and guessed.
+{
+  const gloss = /Words used here[\s\S]{0,1200}/.exec(readme)
+  const missing = ['fieldbook', 'reality', 'terrain', 'trust signal', 'receipts', 'vault']
+    .filter(w => !gloss || !new RegExp(w, 'i').test(gloss[0]))
+  if (!gloss) fail('README must define its invented words once (a "Words used here" line)')
+  else if (missing.length) fail(`README glossary is missing: ${missing.join(', ')}`)
+  else ok('README defines its own vocabulary')
+}
+
 if (!fs.existsSync(path.join(root, 'docs', 'USAGE.md'))) {
   fail('docs/USAGE.md missing')
 } else ok('docs/USAGE.md')

@@ -18,7 +18,7 @@
  *   fde debrief --smart     propose routing from messy notes; --apply commits it
  *   fde prep [label]        grounded walk-in brief from existing .fde/ only
  *   fde doctor              deterministic memory lint (stale signals, gaps)
- *   fde garden [--apply]    propose safe consolidations; apply only with --apply
+ *   fde tidy [--apply]      propose safe consolidations; apply only with --apply (was: garden)
  *   fde ingest …            stage → propose → apply pull sink (.inbox/; never auto-writes .fde/)
  *   fde owner [set …]       who keeps this engagement record
  *   fde receipts <term>     "what did we agree?" - search memory with dates
@@ -2488,18 +2488,18 @@ function cmdGarden(args) {
   // Gardener contract (from Rowboat note_curation): no new facts, no deleted substance,
   // reversible via git when healthy, confirm before apply. Mechanical only - no LLM rewrite.
   if (gitHealth.ok) {
-    console.log('GARDEN (contract: no new facts · no deleted substance · reversible via memory git)')
+    console.log('TIDY (contract: no new facts · no deleted substance · reversible via memory git)')
   } else if (gitHealth.reason === 'broken') {
-    console.log('GARDEN (contract: no new facts · no deleted substance · ⚠ memory git BROKEN — NOT reversible until ledger is repaired)')
+    console.log('TIDY (contract: no new facts · no deleted substance · ⚠ memory git BROKEN — NOT reversible until ledger is repaired)')
   } else {
-    console.log('GARDEN (contract: no new facts · no deleted substance · ⚠ memory not git-versioned — NOT reversible)')
+    console.log('TIDY (contract: no new facts · no deleted substance · ⚠ memory not git-versioned — NOT reversible)')
   }
   console.log(resumeTriage(eng))
   if (!gitHealth.ok) {
     console.log(
       gitHealth.reason === 'broken'
-        ? '\n⚠ ledger is UNVERSIONED (corrupt .git). Repair before trusting garden apply: mv .fde/.git .fde/.git.broken && run any fde write to re-init.'
-        : '\n⚠ no memory git — garden apply cannot create a reversible commit until the ledger exists.'
+        ? '\n⚠ ledger is UNVERSIONED (corrupt .git). Repair before trusting tidy apply: mv .fde/.git .fde/.git.broken && run any fde write to re-init.'
+        : '\n⚠ no memory git — tidy apply cannot create a reversible commit until the ledger exists.'
     )
   }
   const proposals = []
@@ -2539,18 +2539,18 @@ function cmdGarden(args) {
     })
   }
   if (!proposals.length) {
-    console.log('\nNothing to garden.')
+    console.log('\nNothing to tidy.')
     return
   }
   console.log(`\n${proposals.length} proposal(s):`)
   proposals.forEach((p, i) => console.log(`  ${i + 1}. [${p.kind}] ${p.text}`))
   if (!apply) {
-    console.log('\nApply mechanical items only:  fde garden --apply')
+    console.log('\nApply mechanical items only:  fde tidy --apply')
     console.log('Manual items stay yours. Every apply commits to memory git when the ledger is healthy.')
     return
   }
   if (!gitHealth.ok && gitHealth.reason === 'broken') {
-    console.error('refusing garden --apply while memory git is broken - repair the ledger first')
+    console.error('refusing tidy --apply while memory git is broken - repair the ledger first')
     process.exit(1)
   }
   ensureMemoryGit(eng)
@@ -2603,9 +2603,9 @@ function cmdGarden(args) {
     touched.add('context-archive.md')
     console.log(`applied: archived ${p.sessionBlocks.length} old session-end blocks → context-archive.md`)
   }
-  const hash = commitMemory(eng, 'garden', { files: [...touched] })
+  const hash = commitMemory(eng, 'tidy', { files: [...touched] })
   if (!applied) console.log('no mechanical proposals applied (manual items remain)')
-  else console.log(`garden done${hash ? ` @${hash}` : ''}`)
+  else console.log(`tidy done${hash ? ` @${hash}` : ''}`)
 }
 
 // Keep the first open-risk bullet per fingerprint; move later echoes under ## Retired.
@@ -2647,7 +2647,7 @@ function applyRiskDedupe(eng, clusters) {
   const stamp = new Date().toISOString().slice(0, 10)
   const block = retiredLines.map(l => {
     const body = l.trim().replace(/^-\s*/, '')
-    return `- [${stamp}] (garden dedupe) ${body}`
+    return `- [${stamp}] (tidy dedupe) ${body}`
   }).join('\n')
   out = appendUnderSection(out, 'Retired', block)
   withFileLock(p, () => { atomicWriteFile(p, out.endsWith('\n') ? out : out + '\n') })
@@ -3150,14 +3150,13 @@ function printUsage() {
   fde prep [label]         grounded walk-in brief from existing .fde/ only
   fde doctor               lint engagement memory (stale signals, gaps)
   fde redact <term>        preview/remove lines containing a buried term (pass --apply to commit)
-  fde garden [--apply]     propose safe consolidations (contract: no new facts; git-reversible)
+  fde tidy [--apply]       propose safe consolidations (contract: no new facts; git-reversible)
   fde owner [set email]    who keeps this engagement record
   fde receipts <term>      "what did we agree?" with dates
-  fde capture              session-end memory snapshot (hooks use this)
-  fde preserve             pre-compaction context snapshot (hook-internal; hooks use this)
   fde status [--all]       current engagement status (pass --all for full portfolio)
   fde dashboard [--all]    current engagement fieldbook (pass --all for every client)
   fde vault                derived Obsidian vault of every engagement (--current for one, --redacted for a shared screen, --out <dir>)
+  hooks call these; you do not: capture (session-end snapshot), preserve (pre-compaction snapshot)
   env FDEOPS_ENGAGEMENTS_ROOT  override ~/fde-engagements (init/status/dashboard/registry)
   writes require a workspace bind (or FDEOPS_ENGAGEMENT) - folder-name match is read-only
   .fde/ is git-versioned locally for tamper-evident receipts (no remote, no telemetry)
@@ -3176,6 +3175,8 @@ switch (cmd) {
   case 'prep': cmdPrep(args); break
   case 'doctor': cmdDoctor(); break
   case 'redact': cmdRedact(args); break
+  // `garden` was the name through 3.11.x; it keeps working.
+  case 'tidy':
   case 'garden': cmdGarden(args); break
   case 'owner': cmdOwner(args); break
   case 'receipts': cmdReceipts(args); break

@@ -355,7 +355,7 @@ function cmdInstall(opts = {}) {
 // through to the CLI (fde.js reads process.argv itself, so require() is enough).
 const FDE_SUBCOMMANDS = [
   'demo', 'scan', 'resume', 'triage', 'log', 'debrief', 'ingest', 'prep', 'doctor', 'redact',
-  'garden', 'owner', 'receipts', 'capture', 'preserve', 'status', 'dashboard', 'vault', 'help',
+  'tidy', 'garden', 'owner', 'receipts', 'capture', 'preserve', 'status', 'dashboard', 'vault', 'help',
 ]
 
 const INSTALL_SUBCOMMANDS = ['init', 'adapters', 'install']

@@ -6,7 +6,7 @@
 | `adapters/` | Thin per-tool pointers (Codex/`AGENTS.md`, Gemini, Cursor, Copilot, local LLMs) - `node bin/install.js adapters <dir>` |
 | `templates/.fde/` | Core memory templates for `fde resume --init` (phase artifacts are created by phases on demand; `evals.md` is optional) |
 | `examples/` | Fictional walkthroughs with sample `.fde/` files |
-| `bin/fde.js` | Deterministic CLI - scan, resume, triage, log, debrief, ingest, prep, doctor, garden, redact, receipts, capture, preserve, status, dashboard |
+| `bin/fde.js` | Deterministic CLI - scan, resume, triage, log, debrief, ingest, prep, doctor, tidy, redact, receipts, capture, preserve, status, dashboard, vault |
 | `mcp/` | Optional MCP sink (`fdeops-ingest`) + source **recipes** (`mcp/recipes/` — file, granola, notion); source MCPs remain user-configured |
 | `bin/lib/` | Shared memory / trust / render helpers used by the CLI |
 | `bin/check.js` | Structural + install smoke gate (`npm run check`) |

@@ -1,6 +1,6 @@
 {
   "name": "fdeops-ingest-mcp",
-  "version": "3.11.1",
+  "version": "3.12.0",
   "private": true,
   "description": "Thin stdio MCP sink for FDEOps ingest (stage → propose → apply). Zero runtime dependencies.",
   "bin": {

@@ -1,7 +1,7 @@
 {
   "name": "fdeops",
-  "version": "3.11.1",
-  "description": "Field kit for engineers embedded in client work - a real CLI (recon, memory, portfolio), one @fde skill with field judgment on top, and hooks that make it automatic. Claude Code plugin and any agent that loads skills.",
+  "version": "3.12.0",
+  "description": "Engagement memory for AI coding agents. Your agent forgets the client every morning - the sponsor, the promise, who signed off. FDEOps keeps that as dated markdown on your laptop: one @fde skill routing land-to-close methodology, a deterministic local CLI, and hooks that make it automatic. Claude Code plugin and any agent that loads skills.",
   "bin": {
     "fdeops": "bin/install.js",
     "fde": "bin/fde.js"
@@ -30,6 +30,9 @@
     "ai-agent",
     "skills",
     "plugin",
+    "engagement-memory",
+    "agent-memory",
+    "client-context",
     "fde",
     "forward-deployed",
     "forward-deployed-engineer",

@@ -1,8 +1,8 @@
 {
   "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
   "name": "fdeops",
-  "version": "3.11.1",
-  "description": "Engagement fieldbook for Forward Deployed Engineers: per-client memory in local .fde/ files, one @fde skill, land to close methodology. Local-only, no network.",
+  "version": "3.12.0",
+  "description": "Engagement memory for AI coding agents: per-client memory in local .fde/ files, one @fde skill, land to close methodology. Local-only, no network.",
   "author": {
     "name": "Subash Natarajan",
     "url": "https://github.com/suboss87"