docs: document background jobs across prompts, config example, and README

- Extend the injected Background Jobs prompt guidance: system_notifications
  push on completion, collect-only-when-idle wait protocol, and the graph
  LLM-node collect-before-final-turn rule
- Mention the system_notifications push in the agent spawning guidance and
  in the sisyphus/architect wait-protocol text (agent completions push
  notifications too)
- config.example.yaml: max_concurrent_jobs (default 5, 0 = disabled)
- README: features-list entry pointing at the Background-Jobs wiki page

README.md

@@ -51,6 +51,7 @@ Coming from [AIChat](https://github.com/sigoden/aichat)? Follow the [migration g
 * [Skills](https://github.com/Dark-Alex-17/coyote/wiki/Skills): Modular knowledge or capability packs the LLM can load and unload mid-conversation. Multiple skills compose; instructions stack, tools and MCPs union.
 * [Agents](https://github.com/Dark-Alex-17/coyote/wiki/Agents): Leverage AI agents to perform complex tasks and workflows, including sub-agent spawning, teammate messaging, and user interaction tools.
     * [Graph Agents](https://github.com/Dark-Alex-17/coyote/wiki/Graph-Agents): Define an agent as a declarative, YAML-driven workflow. A directed graph of typed nodes (LLM calls, scripts, approvals, user input, RAG retrieval, sub-agent spawns).
+* [Background Jobs](https://github.com/Dark-Alex-17/coyote/wiki/Background-Jobs): Run long tool calls (builds, test suites, slow MCP calls) in the background with the `job__*` tools while the model keeps working — completion arrives as a push notification.
 * [Todo System](https://github.com/Dark-Alex-17/coyote/wiki/TODO-System): Built-in task tracking for improved LLM reliability with smaller models.
 * [Environment Variables](https://github.com/Dark-Alex-17/coyote/wiki/Environment-Variables): Override and customize your Coyote configuration at runtime with environment variables.
 * [Client Configurations](https://github.com/Dark-Alex-17/coyote/wiki/Clients): Configuration instructions for various LLM providers.

assets/agents/architect/config.yaml

@@ -261,7 +261,7 @@ instructions: |
 
   3. **Wait for Sisyphus.** Do not poll `agent__collect` on a running agent — do non-overlapping work
      (e.g. prep the next task's context) or end your response and wait for the completion
-     notification, then `agent__collect`.
+     notification (a `system_notifications` entry on your next tool result), then `agent__collect`.
 
   4. **Verify against the plan (divergence check).** When Sisyphus returns, do NOT trust its
      self-report — get an INDEPENDENT conformance verdict:

assets/agents/sisyphus/config.yaml

@@ -233,7 +233,7 @@ instructions: |
 
   1. Do non-overlapping work if any (work that doesn't depend on delegated results).
   2. If none → **end your response.** Do not call `agent__collect` immediately.
-  3. The system notifies you on completion.
+  3. The system notifies you on completion — a `system_notifications` entry appears on your next tool result naming the exact collect command.
   4. On notification, call `agent__collect` to retrieve results.
 
   ### Anti-duplication rule (BLOCKING)

config.example.yaml

@@ -204,6 +204,7 @@ summary_context_prompt: >        # The text prompt used for including the summar
   'This is a summary of the chat history as a recap: '
 compression_keep_last: 0         # Number of most-recent messages to keep visible after compression (0 = compress all messages)
 max_tool_result_chars: null      # Cap on tool result characters forwarded to the model per call (null = no cap)
+max_concurrent_jobs: 5           # Max background jobs (`job__*` tools) running at once per context (default: 5; 0 disables background jobs entirely)
 
 # ---- Memory ----
 # See the [Memory documentation](https://github.com/Dark-Alex-17/coyote/wiki/Memory) for more information.

src/config/prompts.rs

@@ -112,9 +112,10 @@ pub(in crate::config) const DEFAULT_SPAWN_INSTRUCTIONS: &str = indoc! {"
 
     ### CRITICAL: Never end your turn with pending agents
 
-    Spawned agents do NOT report back on their own. They run in the background until you
+    Spawned agents do NOT deliver their results on their own. When one finishes, a
-    actively reclaim them with `agent__collect` (to get their output) or `agent__cancel`
+    `system_notifications` entry appears on your next tool result naming the exact collect
-    (to discard them). If you spawn agents and then emit a final message without reclaiming
+    command — but the output is only retrieved when you actively reclaim it with `agent__collect`
+    (or discard it with `agent__cancel`). If you spawn agents and then emit a final message without reclaiming
     them, the system will detect the unreclaimed agents and reject the turn-end, injecting
     a reminder forcing you to handle them. After several such reminders, the system will
     auto-cancel them and warn you that work was lost.
@@ -194,10 +195,14 @@ pub(in crate::config) const DEFAULT_JOB_INSTRUCTIONS: &str = indoc! {"
     ## Background Jobs
 
     For long-running tool calls (builds, test suites, slow commands), call `job__start` and keep
-    working instead of blocking. Check progress with `job__check` (sparingly), block on the result
+    working instead of blocking — completion arrives as a `system_notifications` entry on your
-    with `job__collect`, cancel with `job__cancel`, and list jobs with `job__list`. Collect or
+    next tool result. Check progress with `job__check` (sparingly), block on the result with
-    cancel every job you started before ending your turn. Jobs run against a snapshot of the
+    `job__collect` (only when you have nothing else to do), cancel with `job__cancel`, and list
-    current config/environment and do not survive coyote exiting.
+    jobs with `job__list`. Collect or cancel every job you started before ending your turn. In
+    graph LLM nodes, collect or cancel your jobs before ending your final node turn — an
+    uncollected job at node turn-end burns node iterations via the guardrail and can fail the
+    node. Jobs run against a snapshot of the current config/environment and do not survive
+    coyote exiting.
 "
 };