@a5c-ai/atlas-github 5.1.1-staging.d0501abc6dda → 5.1.1-staging.d77cc7fd54ae

package/bin/install-shared.js

@@ -104,7 +104,7 @@ function ensureMarketplaceEntry(marketplacePath, pluginRoot) {
     name: PLUGIN_NAME,
     source: relSource,
     description: "Turn a stated need into a full system design by mining the Atlas knowledge graph.",
-    version: "5.1.1-staging.d0501abc6dda",
+    version: "5.1.1-staging.d77cc7fd54ae",
     author: { name: "a5c.ai" },
   };
   if (idx >= 0) marketplace.plugins[idx] = entry;

package/commands/collect-nuances.md

@@ -1,10 +1,18 @@
 ---
-description: Collect nuances — gather domain-specific edge cases, constraints, and gotchas from the Atlas graph.
-argument-hint: The domain/system to collect nuances for.
+description: Collect the REAL constraints/gotchas of your scanned systems (IaC drift, orphaned resources, RBAC quirks, region splits), each cited; Atlas-graph comparison secondary.
+argument-hint: The sources to inspect for nuances (e.g. "our azure account and this repo into C:/work/atlas").
 allowed-tools: Read, Grep, Write, Task, Bash, Edit, Glob, AskUserQuestion, TodoWrite, Skill, mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_wiki_page
 ---
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-collect-nuances`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-collect-nuances.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-collect-nuances --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-collect-nuances.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans (Bash `az` + git/fs) of the user's REAL systems to collect actual constraints/gotchas/edge-cases — IaC-as-survey drift, orphaned cloud resources, RBAC/visibility quirks, region splits, tenant/descriptor mismatches, externally-deployed systems with no in-repo pipeline — each cited to its real resource id / RG / file path (or a concrete missing-thing observation). The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent constraints; only inspect the sources named in the arguments.
 
 User arguments for this command:
 

package/commands/discover.md

@@ -1,10 +1,18 @@
 ---
-description: Interactive systems discovery — turn a stated need into a full system design via the Atlas graph.
-argument-hint: The system/need to design.
+description: Scan your real systems (Azure + repos + dirs) and map them into a layered, source-cited atlas; enrich against the Atlas graph (secondary).
+argument-hint: The sources to scan (e.g. "our azure account, this repo and C:/work/ into C:/work/atlas").
 allowed-tools: Read, Grep, Write, Task, Bash, Edit, Glob, AskUserQuestion, TodoWrite, Skill, mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_wiki_page
 ---
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-systems-discovery`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-systems-discovery.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-systems-discovery --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-systems-discovery.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans of the user's REAL systems — Azure via `az` (Bash), git repos, and local directories — and synthesizes a cross-linked, source-cited atlas where every item cites a real resource id / RG / file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY enrichment/comparison, never as the primary content. Never invent resource ids or file paths; only scan the sources named in the arguments.
 
 User arguments for this command:
 

package/commands/mine-data.md

@@ -1,10 +1,18 @@
 ---
-description: Data mining — discover the data models/entities a domain requires from the Atlas graph.
-argument-hint: The domain or goal to mine data for.
+description: Mine the REAL data stores/models in your sources (cloud DBs via az, schemas/migrations, package types), each cited; Atlas-graph comparison secondary.
+argument-hint: The sources to mine for data (e.g. "our azure account and this repo into C:/work/atlas").
 allowed-tools: Read, Grep, Write, Task, Bash, Edit, Glob, AskUserQuestion, TodoWrite, Skill, mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_wiki_page
 ---
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-data-mining`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-data-mining.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-data-mining --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-data-mining.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans to mine the REAL data — cloud databases/storage/AI-Search via `az` (Bash), and schemas/migrations/ORM models/package types in the repos/dirs — binding models to the stores that back them, each cited to its real resource id / file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent stores or models; only mine the sources named in the arguments. Never read secret values.
 
 User arguments for this command:
 

package/commands/mine-processes.md

@@ -1,10 +1,18 @@
 ---
-description: Process mining — discover the processes/workflows a domain requires from the Atlas graph.
-argument-hint: The domain or goal to mine processes for.
+description: Mine the REAL processes in your sources (CI/CD, npm scripts, IaC, Dockerfiles, .a5c, cron), each cited to its file; Atlas-graph comparison secondary.
+argument-hint: The sources to mine (e.g. "this repo and C:/work/ into C:/work/atlas").
 allowed-tools: Read, Grep, Write, Task, Bash, Edit, Glob, AskUserQuestion, TodoWrite, Skill, mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_wiki_page
 ---
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-process-mining`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-process-mining.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-process-mining --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-process-mining.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans (Bash) of the REAL repos/dirs/cloud automation and mines the processes that actually exist — `.github/workflows`, npm scripts, terraform/bicep/helm/k8s, Dockerfiles, babysitter `.a5c` processes, and cron/scheduled jobs — each cited to its real file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent files or processes; only mine the sources named in the arguments.
 
 User arguments for this command:
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@a5c-ai/atlas-github",
-  "version": "5.1.1-staging.d0501abc6dda",
+  "version": "5.1.1-staging.d77cc7fd54ae",
   "description": "Turn a stated need into a full system design by mining the Atlas knowledge graph.",
   "scripts": {
     "deploy": "npm publish --access public",
@@ -34,7 +34,7 @@
     "access": "public"
   },
   "dependencies": {
-    "@a5c-ai/babysitter-sdk": "5.1.1-staging.d0501abc6dda"
+    "@a5c-ai/babysitter-sdk": "5.1.1-staging.d77cc7fd54ae"
   },
   "repository": {
     "type": "git",

package/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "atlas",
-  "version": "5.1.1-staging.d0501abc6dda",
+  "version": "5.1.1-staging.d77cc7fd54ae",
   "description": "Turn a stated need into a full system design by mining the Atlas knowledge graph.",
   "author": {
     "name": "a5c.ai"

package/skills/atlas/SKILL.md

@@ -1,56 +1,74 @@
 ---
 name: atlas
 description: >
-  Turn a stated need into a full system design by mining the Atlas knowledge
-  graph. Use this skill when asked to design a system from a goal, discover the
-  components/processes/data a domain needs, mine processes or data models, or
-  collect domain nuances. (atlas, design a system, what does X need, system
-  discovery, process mining, data mining, collect nuances, blueprint from a need,
-  graph-driven design)
+  Atlas turns your STATED NEED into a real systems atlas by SCANNING your actual
+  sources (Azure via `az`, git repos, local dirs) and process/data mining them,
+  THEN enriching against the Atlas knowledge graph. Use this skill when asked to
+  inventory/map your real systems, scan your cloud + repos + directories, mine the
+  real processes or data they contain, or collect their real constraints/gotchas.
+  (atlas, scan my systems, inventory our azure account, map my repos, real systems
+  atlas, process mining, data mining, collect nuances, system discovery)
 allowed-tools: Read, Grep, Glob, Write, Edit, Task, Bash, AskUserQuestion, TodoWrite, Skill, mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_kind, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_stats, mcp__atlas__atlas_public_wiki_page
 version: 0.1.0
 ---
 
 # atlas
 
-This skill turns a stated need into an entire system design by mining the Atlas
-knowledge graph. It is the brain of the `atlas` plugin. For non-trivial runs it
-delegates orchestration to `babysitter:babysit` using an atlas-specific `.a5c`
-process; for simple lookups it queries the graph directly.
+This skill turns a stated need into a **real systems atlas** by SCANNING your
+actual sources — Azure subscriptions (via read-only `az`), git repos, and local
+directories — and process/data mining them, THEN enriching the result against the
+Atlas knowledge graph. It is the brain of the `atlas` plugin. The scan is
+PRIMARY; the graph is SECONDARY. For non-trivial runs it delegates orchestration
+to `babysitter:babysit` using an atlas-specific `.a5c` process; for simple
+lookups it queries the graph directly.
 
-## 1. What Atlas is
+## 1. Scan-first, graph-second
 
-Atlas is a knowledge graph of agents, processes, data models, capabilities,
-workflows, clusters, and wiki pages. You reach it through the
-`mcp__atlas__atlas_public_*` MCP tools, which this plugin wires natively into
-your harness (see the plugin README "Atlas MCP" section; the server URL is
-overridable via `ATLAS_MCP_URL`). Never assume the graph's shape — discover it
-through the tools.
+The output you want is an evidence-backed inventory of **your** systems — e.g.
+`azure-inventory.json` (every real resource id + RG from `az`),
+`workspace-inventory.json` (real repo/dir scan), `processes.json` (real mined
+CI/CD/IaC/.a5c processes), and a cross-linked `SYSTEMS-ATLAS.md`. Every item must
+cite its REAL source. Generic catalog nodes are NOT the deliverable.
+
+- **Primary — scan the user's real sources.** Use `Bash` to run READ-ONLY scans:
+  `az` (account/group/resource list + per-service list/show) for Azure;
+  `git` + filesystem (`Read`/`Glob`) for repos and directories. NEVER invent
+  resource ids, regions, SKUs, or file paths — if you didn't observe it in real
+  output, it does not go in the atlas. Only scan the sources named in the need
+  (scoping, not a fallback).
+- **Secondary — the Atlas knowledge graph.** Atlas is a knowledge graph of
+  agents, processes, data models, capabilities, workflows, and wiki pages reached
+  through the `mcp__atlas__atlas_public_*` MCP tools (server URL overridable via
+  `ATLAS_MCP_URL`). Use it ONLY to add best-practice / comparison context for the
+  real systems you found — never as the primary content, never to pad the atlas
+  with generic nodes. See the `atlas-graph-query` skill for the tool surface.
 
 ## 2. When to use
 
 | Trigger phrase | Command |
 |----------------|---------|
-| design a system, blueprint from a need, graph-driven design | `/atlas:discover` |
-| process mining, what processes does X need | `/atlas:mine-processes` |
-| data mining, what data does X need | `/atlas:mine-data` |
-| collect nuances, edge cases/constraints for X | `/atlas:collect-nuances` |
+| scan/inventory my real systems (azure + repos + dirs), map them | `/atlas:discover` |
+| mine the real processes in my repos/cloud (CI/CD, IaC, .a5c, cron) | `/atlas:mine-processes` |
+| mine the real data stores/models in my cloud + repos | `/atlas:mine-data` |
+| collect the real constraints/gotchas of my scanned systems | `/atlas:collect-nuances` |
 
-## 3. The need → design pipeline (core method)
+## 3. The need → real atlas pipeline (core method)
 
-1. **Frame the need** — restate the goal as a domain + concrete outcomes. If the
-   request is ambiguous, run a short interview (`AskUserQuestion`). Per repo
-   policy, interview ONLY when requirements are genuinely unclear.
-2. **Locate anchors** — `atlas_public_search(q=<need terms>)` to find seed nodes;
-   `atlas_public_clusters` / `atlas_public_kinds` to scope the domain.
-3. **Expand the graph** — `atlas_public_neighbors(id, depth, edges, kinds)` from
-   the anchors to gather the system's parts: components, workflows, processes,
-   data models, capabilities.
-4. **Read detail** — `atlas_public_record(id, expandNeighbors)` for fields and
-   edges; `atlas_public_wiki_page` for narrative context.
-5. **Synthesize the design** — assemble the discovered nodes into a layered
-   system design (components, processes, data, integrations, nuances) and write a
-   design artifact under `.a5c/atlas/<run>/design.md` plus a machine mirror.
+1. **Parse sources** — interpret the stated need into concrete SOURCES: Azure
+   subscription(s), git repos, local directories, URLs, plus the output dir. If
+   the sources are genuinely ambiguous, run a short interview
+   (`AskUserQuestion`). Per repo policy, interview ONLY when truly unclear.
+2. **Scan cloud (primary)** — for each Azure source, run read-only `az` and write
+   a real cloud inventory citing resource ids/RGs. Skip cleanly (record a reason)
+   if no cloud source is in scope — only scan what's named.
+3. **Scan local (primary)** — for each repo/dir, scan the filesystem + git
+   (structure, submodules, manifests, languages, services, IaC) and write a real
+   inventory citing real paths.
+4. **Enrich (secondary)** — map the discovered real systems against the Atlas
+   graph for comparison context. Clearly secondary; never the headline.
+5. **Synthesize** — assemble a real, cross-linked layered atlas (components /
+   processes / data / integrations / nuances) where EVERY item cites its real
+   source, like `SYSTEMS-ATLAS.md`, plus a machine mirror.
 6. **Converge (TDD)** — each phase asserts its own checkable outputs before
    proceeding (see the atlas processes), iterating until the assertions pass.
 
@@ -68,10 +86,18 @@ Do not hand-roll orchestration when a process exists.
 
 ## 5. Guardrails
 
-- No fallbacks (repo rule). If you find yourself writing a fallback, stop and fix
-  the root cause.
-- Never invent graph node ids — only reference ids returned by Atlas tools.
-- Keep breakpoints sparse; use them only when user input is genuinely critical or
+- No fallbacks (repo rule). Skipping an out-of-scope source class (e.g. no cloud
+  named) is correct scoping and must be recorded with a reason — it is NOT a
+  silent fallback to the public graph. If you find yourself writing a real
+  fallback, stop and fix the root cause.
+- Scan-first: every system/item in the atlas MUST cite a REAL source (an `az`
+  resource id / RG, or a file path). Never invent resource ids, regions, SKUs, or
+  file paths. Never invent graph node ids either — only reference ids returned by
+  the Atlas tools, and keep graph content strictly secondary.
+- Read-only scanning only: `az` read verbs, `git` status/remote/log, filesystem
+  reads. Never run mutating cloud/git/fs commands and never read secret values.
+- Keep breakpoints sparse; use them only when the sources to scan are genuinely
   ambiguous.
-- Do not emit `kind: 'shell'` subtasks unless the user explicitly asks for a
-  shell-oriented workflow.
+- The real scanning is done BY the agent via its `Bash` tool inside the agent
+  task prompt. Do not emit `kind: 'shell'` subtasks unless the user explicitly
+  asks for a shell-oriented workflow.

package/skills/atlas-graph-query/SKILL.md

@@ -1,10 +1,12 @@
 ---
 name: atlas-graph-query
 description: >
-  Reference for querying the Atlas knowledge graph through its MCP tools. Use
-  when you need to look up nodes, edges, kinds, clusters, stats, or wiki pages in
-  Atlas. (atlas graph, query atlas, atlas mcp, search the graph, graph neighbors,
-  atlas record, atlas kinds)
+  Reference for querying the Atlas knowledge graph through its MCP tools — the
+  SECONDARY enrichment/comparison layer that adds best-practice context to systems
+  you have ALREADY scanned from your real sources (`az`, repos, dirs). Use when
+  you need to look up nodes, edges, kinds, clusters, stats, or wiki pages in Atlas
+  to compare against your real inventory. (atlas graph, query atlas, atlas mcp,
+  search the graph, graph neighbors, atlas record, atlas kinds, enrichment layer)
 allowed-tools: mcp__atlas__atlas_public_search, mcp__atlas__atlas_public_record, mcp__atlas__atlas_public_neighbors, mcp__atlas__atlas_public_kinds, mcp__atlas__atlas_public_kind, mcp__atlas__atlas_public_edge_kinds, mcp__atlas__atlas_public_edge_kind, mcp__atlas__atlas_public_clusters, mcp__atlas__atlas_public_stats, mcp__atlas__atlas_public_wiki_page
 version: 0.1.0
 ---
@@ -16,6 +18,14 @@ A thin reference for the Atlas knowledge-graph MCP tool surface so any agent
 server URL is wired natively by the `atlas` plugin and is overridable via
 `ATLAS_MCP_URL`. Never invent node ids — only use ids returned by these tools.
 
+> **Position: this is the SECONDARY / enrichment layer.** The `atlas` plugin is
+> scan-first — it inventories your REAL systems by scanning your actual sources
+> (Azure via read-only `az`, git repos, local directories) and process/data
+> mining them. The graph queries below are used ONLY to add best-practice /
+> comparison context to those already-discovered real systems. Do NOT use the
+> graph as the primary content, and never pad a real inventory with generic
+> catalog nodes. Tie every graph lookup back to a real scanned system.
+
 ## Tools
 
 ### `mcp__atlas__atlas_public_search`

package/skills/collect-nuances/SKILL.md

@@ -1,11 +1,19 @@
 ---
 name: collect-nuances
-description: Collect nuances — gather domain-specific edge cases, constraints, and gotchas from the Atlas graph.
+description: Collect the REAL constraints/gotchas of your scanned systems (IaC drift, orphaned resources, RBAC quirks, region splits), each cited; Atlas-graph comparison secondary.
 ---
 
 # collect-nuances
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-collect-nuances`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-collect-nuances.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-collect-nuances --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-collect-nuances.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans (Bash `az` + git/fs) of the user's REAL systems to collect actual constraints/gotchas/edge-cases — IaC-as-survey drift, orphaned cloud resources, RBAC/visibility quirks, region splits, tenant/descriptor mismatches, externally-deployed systems with no in-repo pipeline — each cited to its real resource id / RG / file path (or a concrete missing-thing observation). The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent constraints; only inspect the sources named in the arguments.
 
 User arguments for this command:
 

package/skills/discover/SKILL.md

@@ -1,11 +1,19 @@
 ---
 name: discover
-description: Interactive systems discovery — turn a stated need into a full system design via the Atlas graph.
+description: Scan your real systems (Azure + repos + dirs) and map them into a layered, source-cited atlas; enrich against the Atlas graph (secondary).
 ---
 
 # discover
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-systems-discovery`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-systems-discovery.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-systems-discovery --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-systems-discovery.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans of the user's REAL systems — Azure via `az` (Bash), git repos, and local directories — and synthesizes a cross-linked, source-cited atlas where every item cites a real resource id / RG / file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY enrichment/comparison, never as the primary content. Never invent resource ids or file paths; only scan the sources named in the arguments.
 
 User arguments for this command:
 

package/skills/mine-data/SKILL.md

@@ -1,11 +1,19 @@
 ---
 name: mine-data
-description: Data mining — discover the data models/entities a domain requires from the Atlas graph.
+description: Mine the REAL data stores/models in your sources (cloud DBs via az, schemas/migrations, package types), each cited; Atlas-graph comparison secondary.
 ---
 
 # mine-data
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-data-mining`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-data-mining.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-data-mining --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-data-mining.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans to mine the REAL data — cloud databases/storage/AI-Search via `az` (Bash), and schemas/migrations/ORM models/package types in the repos/dirs — binding models to the stores that back them, each cited to its real resource id / file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent stores or models; only mine the sources named in the arguments. Never read secret values.
 
 User arguments for this command:
 

package/skills/mine-processes/SKILL.md

@@ -1,11 +1,19 @@
 ---
 name: mine-processes
-description: Process mining — discover the processes/workflows a domain requires from the Atlas graph.
+description: Mine the REAL processes in your sources (CI/CD, npm scripts, IaC, Dockerfiles, .a5c, cron), each cited to its file; Atlas-graph comparison secondary.
 ---
 
 # mine-processes
 
-Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md) to run the atlas process `atlas-process-mining`. Pass the user arguments below as the run's stated need. Then continue executing the returned instructions in this same turn; do not stop after the Skill tool returns. Use the atlas skill and the Atlas MCP tools (mcp__atlas__atlas_public_*) for all graph queries.
+Invoke the babysitter:babysit skill (using the Skill tool) and follow its instructions (SKILL.md). Run the PRE-AUTHORED atlas process bundled with this plugin — do NOT interview the user or author a new process. Create the run directly from this bundled entry and iterate it to completion:
+
+```
+${COPILOT_PLUGIN_DIR}/processes/atlas-process-mining.mjs#process
+```
+
+(e.g. `babysitter run:create --process-id atlas-process-mining --entry "${COPILOT_PLUGIN_DIR}/processes/atlas-process-mining.mjs#process" --harness <this-harness>`, then iterate.) Pass the user arguments below as the run's stated need / process inputs. Continue executing in this same turn; do not stop after the Skill tool returns.
+
+This process is SCAN-FIRST: it parses the stated sources, then runs READ-ONLY scans (Bash) of the REAL repos/dirs/cloud automation and mines the processes that actually exist — `.github/workflows`, npm scripts, terraform/bicep/helm/k8s, Dockerfiles, babysitter `.a5c` processes, and cron/scheduled jobs — each cited to its real file path. The Atlas knowledge graph (mcp__atlas__atlas_public_*) is used only as SECONDARY comparison. Never invent files or processes; only mine the sources named in the arguments.
 
 User arguments for this command:
 

package/versions.json

@@ -1,4 +1,4 @@
 {
-  "sdkVersion": "5.1.1-staging.d0501abc6dda",
-  "extensionVersion": "5.1.1-staging.d0501abc6dda"
+  "sdkVersion": "5.1.1-staging.d77cc7fd54ae",
+  "extensionVersion": "5.1.1-staging.d77cc7fd54ae"
 }