test(suite): replace plan-label references with self-contained wording

Rust test comments/assert messages and deep-research script docstrings
cited numbered requirement labels (R3/R5) and task labels (TASK-016,
TASK-021) that are defined only in an external plan document. Reword
each to say what it means on its own: fail-safe crash guards, the
validator's fallback-capture warning, task-queue anchors.

assets/agents/deep-research/README.md

@@ -263,7 +263,7 @@ pass — every fault is visible to the reviewer or the caller — and some
 end node always renders its sentinel, so the graph degrades instead of
 dying.
 
-All 11 scripts also carry a top-level crash guard (R3): a script that
+All 11 scripts also carry a top-level crash guard: a script that
 throws records a fault and fails forward where the flow allows it (a
 crashed `reflexion_gate` proceeds to `synthesize`; a crashed
 `verify_sources` reports that sources were NOT checked) instead of

assets/agents/deep-research/scripts/bootstrap_research.py

@@ -7,7 +7,7 @@ the RAG knowledge lookup) as a single super-step. The validator requires
 declared parallel-branch script outputs, so we emit an empty JSON object
 explicitly here.
 
-Fail-safe (R3): the node reads no state and computes nothing, so its
+Fail-safe: the node reads no state and computes nothing, so its
 happy-path output IS the sane degraded output — the guard emits the same
 empty object on any crash. No PIPELINE-FAULT is recorded: appending one
 would require reading state (the very thing that crashed) or clobbering

assets/agents/deep-research/scripts/combine_findings.py

@@ -14,7 +14,7 @@ surface their collected output_key). This script lifts such findings into
 `pipeline_faults` (deduplicated, since a reflexion or feedback loop can
 re-run the map) so dead lanes surface in the report's Pipeline notes.
 
-Fail-safe (R3): on a crash (e.g. malformed GRAPH_STATE), `findings`
+Fail-safe: on a crash (e.g. malformed GRAPH_STATE), `findings`
 becomes the fault text itself and a PIPELINE-FAULT entry is recorded —
 the pipeline proceeds visibly degraded, and the human approval gate is
 the backstop. Existing faults are preserved best-effort.

assets/agents/deep-research/scripts/incorporate_feedback.py

@@ -11,7 +11,7 @@ budget.
 
 Routing (`_next`): always research_each_question.
 
-Fail-safe (R3): a crash here means the user's feedback text cannot be
+Fail-safe: a crash here means the user's feedback text cannot be
 recovered, but their INTENT — another research pass — is unambiguous
 (they chose neither accept nor reject). The degraded route therefore
 still loops to `research_each_question`, with a generic feedback note

assets/agents/deep-research/scripts/parse_request.py

@@ -10,7 +10,7 @@ Routing (`_next`):
   - prompt present -> (no _next; static next: plan)
   - prompt empty   -> ask_topic
 
-Fail-safe (R3): a crash here (e.g. malformed GRAPH_STATE) means the
+Fail-safe: a crash here (e.g. malformed GRAPH_STATE) means the
 caller's prompt cannot be read at all, so proceeding silently is
 meaningless — the degraded route is `ask_topic`, which asks the user for
 the topic directly, plus a PIPELINE-FAULT entry so the loss is visible in

assets/agents/deep-research/scripts/reflexion_gate.py

@@ -16,7 +16,7 @@ Reflexion is a best-effort quality booster, not a hard gate: once the
 budget is spent the workflow proceeds anyway, and the human approval
 step is the final backstop.
 
-Fail-safe (R3): a crash here fails FORWARD to `synthesize` — the same
+Fail-safe: a crash here fails FORWARD to `synthesize` — the same
 direction as the existing malformed-critique PASS default — with a
 PIPELINE-FAULT entry recorded (best-effort preserving existing faults),
 so a broken gate costs the run its automated critique routing, never the

assets/agents/deep-research/scripts/verify_sources.py

@@ -15,7 +15,7 @@ rendered section when `pipeline_faults` is non-empty and the empty
 string otherwise, keeping the happy-path `{{report}}{{pipeline_notes}}`
 output byte-identical to the pre-hardening `{{report}}`.
 
-Fail-safe (R3): per-URL probes were already guarded; the top level now
+Fail-safe: per-URL probes were already guarded; the top level now
 is too. On a crash the summary says sources were NOT checked, a
 PIPELINE-FAULT is recorded (best-effort preserving existing faults), and
 pipeline_notes still renders — the approval gate remains the backstop.

src/config/request_context.rs

@@ -8181,7 +8181,7 @@ mod tests {
         ] {
             assert!(
                 architect.contains(anchor),
-                "architect config lost TASK-021 anchor: {anchor:?}"
+                "architect config lost task-queue mirroring anchor: {anchor:?}"
             );
         }
 
@@ -8199,7 +8199,7 @@ mod tests {
         ] {
             assert!(
                 sisyphus.contains(anchor),
-                "sisyphus config lost TASK-021 anchor: {anchor:?}"
+                "sisyphus config lost task-queue chain anchor: {anchor:?}"
             );
         }
 
@@ -8214,7 +8214,7 @@ mod tests {
         ] {
             assert!(
                 spawn.contains(anchor),
-                "DEFAULT_SPAWN_INSTRUCTIONS lost TASK-021 anchor: {anchor:?}"
+                "DEFAULT_SPAWN_INSTRUCTIONS lost task-queue semantics anchor: {anchor:?}"
             );
         }
         assert!(
@@ -8492,7 +8492,7 @@ mod tests {
         );
     }
 
-    // ---- adversary suite-script regression tests (R5) ----
+    // ---- adversary suite-script regression tests ----
     //
     // Fault paths of the adversary's verdict/gate scripts, exercised by
     // invoking `python3 <script>` with a synthetic GRAPH_STATE env — the same
@@ -9427,7 +9427,7 @@ mod tests {
         assert_eq!(holistic.max_iterations, 20);
     }
 
-    // ---- review-gauntlet suite-script regression tests (R5) ----
+    // ---- review-gauntlet suite-script regression tests ----
     //
     // Degradation paths of the gauntlet's lane/builder/gate scripts,
     // exercised the same way as the adversary suite above: `python3 <script>`
@@ -9914,7 +9914,7 @@ mod tests {
         );
     }
 
-    // ---- code-reviewer suite-script regression tests (R5) ----
+    // ---- code-reviewer suite-script regression tests ----
     //
     // Fail-closed fault paths of the code-reviewer's verdict/fault/render
     // scripts, exercised the same way as the suites above: `python3 <script>`
@@ -10503,7 +10503,7 @@ mod tests {
         );
     }
 
-    // ---- finding-verifier suite-script regression tests (R5) ----
+    // ---- finding-verifier suite-script regression tests ----
     //
     // Fail-closed fault paths of the finding-verifier's marker scripts,
     // exercised the same way as the suites above: `python3 <script>` with a
@@ -10738,7 +10738,7 @@ mod tests {
         );
     }
 
-    // ---- step-runner fault-wiring regression tests (R5) ----
+    // ---- step-runner fault-wiring regression tests ----
     //
     // route_review.sh is a bash script node; the graph's script executor runs
     // `.sh` scripts through bash with the same GRAPH_STATE env contract, so
@@ -10899,7 +10899,7 @@ mod tests {
         if skip_step_runner_bash_harness() {
             return;
         }
-        // R3: a fault-noting script must never itself kill the pipeline —
+        // A fault-noting script must never itself kill the pipeline —
         // the harness asserts exit 0 + JSON stdout even for a bare state.
         let out = run_step_runner_script("note_llm_fault.sh", &json!({}));
         assert_eq!(out, json!({"fault_note": ""}));
@@ -10949,8 +10949,9 @@ mod tests {
         assert_eq!(review.fallback.as_deref(), Some("write_handoff"));
 
         // The llm nodes with fallbacks capture their failure text so the
-        // fallback path can see why it was reached (the TASK-016 warning —
-        // this is what shrank the step-runner warning baseline to zero).
+        // fallback path can see why it was reached (the validator's
+        // fallback-capture warning — this is what shrank the step-runner
+        // warning baseline to zero).
         // orient and write_handoff route their fallbacks through
         // note_llm_fault, which distills the capture into fault_note;
         // edge_case_sweep still falls back to write_handoff, which renders
@@ -11037,9 +11038,9 @@ mod tests {
         }
     }
 
-    // ---- deep-research suite-script regression tests (R5) ----
+    // ---- deep-research suite-script regression tests ----
     //
-    // The deep-research graph's R3 guards and fault markers, exercised by
+    // The deep-research graph's crash guards and fault markers, exercised by
     // invoking `python3 <script>` with a synthetic GRAPH_STATE env. The raw
     // runner feeds deliberately malformed JSON to prove every guarded script
     // emits a sane degraded output instead of crashing the node.
@@ -11077,7 +11078,7 @@ mod tests {
             eprintln!("skipping: python3 not available");
             return;
         }
-        // R3: every guarded script and fault marker must exit 0 with JSON
+        // Fail-safe: every guarded script and fault marker must exit 0 with JSON
         // stdout even when GRAPH_STATE is not JSON at all (the runner
         // asserts both); per-script degraded shapes are pinned below.
         for script in [
@@ -11516,8 +11517,8 @@ mod tests {
             .unwrap();
 
         // plan: fail-closed. Its output_schema + fallback combination requires
-        // a state_updates capture (the TASK-016 warning) so plan_fault can see
-        // why it was reached.
+        // a state_updates capture (the validator's fallback-capture warning) so
+        // plan_fault can see why it was reached.
         let NodeType::Llm(plan) = &graph.get_node("plan").unwrap().node_type else {
             panic!("plan must be an llm node")
         };