diff --git a/.changeset/sturdy-koalas-zip.md b/.changeset/sturdy-koalas-zip.md
new file mode 100644
index 000000000..7fc0bc106
--- /dev/null
+++ b/.changeset/sturdy-koalas-zip.md
@@ -0,0 +1,5 @@
+---
+type: Fixed
+pr: 3449
+---
+planning-config.md documented "light" as an allowed workflow.code_review_depth value, but config-set only accepts quick/standard/deep — the reference now matches the validator, pinned by a doc↔capability-registry parity test. The agent_skills row now also documents the array-of-strings form for assigning multiple skill sets to one agent type.
diff --git a/gsd-core/references/planning-config.md b/gsd-core/references/planning-config.md
index 9beedde7f..8b48b3e0f 100644
--- a/gsd-core/references/planning-config.md
+++ b/gsd-core/references/planning-config.md
@@ -275,7 +275,7 @@ Set via `workflow.*` namespace in config.json (e.g., `"workflow": { "research":
 | `workflow.code_review_command` | string\|null | `null` | Any shell command | External code-review command integrated into `/gsd:ship`. The diff is piped to the command via stdin; the command must output JSON with a `verdict` field (`"APPROVED"` or `"REVISE"`). Non-zero exit or `"REVISE"` verdict blocks the ship workflow. When unset, the built-in review flow runs. Example: `my-review-tool --review`. |
 | `workflow.inline_plan_threshold` | number | `2` | `0`–`10` | Plans with ≤N tasks execute inline instead of spawning a subagent |
 | `workflow.code_review` | boolean | `true` | `true`, `false` | Enable built-in code review step in the ship workflow |
-| `workflow.code_review_depth` | string | `"standard"` | `"light"`, `"standard"`, `"deep"` | Depth level for code review analysis in the ship workflow |
+| `workflow.code_review_depth` | string | `"standard"` | `"quick"`, `"standard"`, `"deep"` | Depth level for code review analysis in the ship workflow |
 | `workflow._auto_chain_active` | boolean | `false` | `true`, `false` | Internal: tracks whether autonomous chaining is active |
 | `workflow.security_enforcement` | boolean | `true` | `true`, `false` | Enable threat-model-anchored security verification via `/gsd:secure-phase`. When `false`, security checks are skipped entirely |
 | `workflow.security_asvs_level` | number | `1` | `1`, `2`, `3` | OWASP ASVS verification level. Level 1 = opportunistic, Level 2 = standard, Level 3 = comprehensive. Scales both planner threat-disposition rigor (which threats must be mitigated vs. accepted) and auditor verification depth (grep-level → boundary-placement check → full data-flow trace). See `gsd-core/references/security-asvs-levels.md`. |
@@ -362,7 +362,7 @@ Set via `manager.*` namespace (e.g., `"manager": { "flags": { "discuss": "--auto
 |-----|------|---------|----------------|-------------|
 | `parallelization` | boolean\|object | `true` | `true`, `false`, `{ "enabled": true }` | Enable parallel wave execution; object form allows additional sub-keys |
 | `model_overrides` | object\|null | `null` | `{ "<agent-type>": "<model-id>" }` | Override model selection per agent type |
-| `agent_skills` | object | `{}` | `{ "<agent-type>": "<skill-set>" }` | Assign skill sets to specific agent types |
+| `agent_skills` | object | `{}` | `{ "<agent-type>": "<skill-set>" }` or `{ "<agent-type>": ["<skill-set>", "<skill-set>", ...] }` | Assign skill sets to specific agent types. Each value is a single skill-set path (string) or an array of skill-set paths — the array form assigns multiple skill sets to one agent type. Paths cannot be comma-joined into one string; each path must be its own array element |
 | `sub_repos` | array | `[]` | Array of relative path strings | Child directories with independent `.git` repos (auto-detected) |
 
 ### Planning Fields
diff --git a/tests/config-field-docs.test.cjs b/tests/config-field-docs.test.cjs
index 9ff819c05..2630841ec 100644
--- a/tests/config-field-docs.test.cjs
+++ b/tests/config-field-docs.test.cjs
@@ -238,6 +238,82 @@ describe('config-field-docs', () => {
   });
 });
 
+// ─── Capability-enum doc parity (#3303) ─────────────────────────────────────
+
+describe('capability-enum doc parity (#3303)', () => {
+  const CAPABILITY_PATH = path.join(
+    __dirname, '..', 'capabilities', 'code-review', 'capability.json'
+  );
+
+  let docContent;
+  let capability;
+
+  before(() => {
+    docContent = fs.readFileSync(REFERENCE_PATH, 'utf-8');
+    capability = JSON.parse(fs.readFileSync(CAPABILITY_PATH, 'utf-8'));
+  });
+
+  /**
+   * Extract a single Complete-Field-Reference table row by key and split it into
+   * trimmed cells: ['', '`key`', type, default, allowedValues, description, ''].
+   * Row-scoping keeps unrelated wording elsewhere in the doc from satisfying
+   * (or breaking) the parity assertions.
+   */
+  function docRow(key) {
+    const line = docContent
+      .split(/\r?\n/)
+      .find(l => l.startsWith(`| \`${key}\` |`));
+    assert.ok(line, `planning-config.md must have a table row for \`${key}\``);
+    return line.split('|').map(c => c.trim());
+  }
+
+  test('workflow.code_review_depth allowed values match the capability registry (#3303)', () => {
+    const spec = capability.config && capability.config['workflow.code_review_depth'];
+    assert.ok(
+      spec && Array.isArray(spec.values),
+      'capabilities/code-review/capability.json must declare workflow.code_review_depth values'
+    );
+
+    const allowedCell = docRow('workflow.code_review_depth')[4];
+    const docValues = [...allowedCell.matchAll(/"([^"]+)"/g)].map(m => m[1]);
+    assert.ok(
+      docValues.length > 0,
+      `workflow.code_review_depth Allowed-Values cell must list quoted values, got: ${allowedCell}`
+    );
+    assert.deepStrictEqual(
+      [...docValues].sort(),
+      [...spec.values].sort(),
+      `planning-config.md workflow.code_review_depth allowed values (${docValues.join(', ')}) ` +
+        `must exactly match what config-set accepts per capability.json (${spec.values.join(', ')})`
+    );
+  });
+
+  test('workflow.code_review_depth default matches the capability registry (#3303)', () => {
+    const spec = capability.config && capability.config['workflow.code_review_depth'];
+    assert.ok(spec, 'capability.json must declare workflow.code_review_depth');
+    const defaultCell = docRow('workflow.code_review_depth')[3];
+    const [docDefault] = [...defaultCell.matchAll(/"([^"]+)"/g)].map(m => m[1]);
+    assert.strictEqual(
+      docDefault,
+      spec.default,
+      `planning-config.md workflow.code_review_depth default must match capability.json default (${spec.default})`
+    );
+  });
+
+  test('agent_skills row documents the array-of-strings form (#3303)', () => {
+    const cells = docRow('agent_skills');
+    const allowedCell = cells[4];
+    assert.ok(
+      /\[\s*"<skill-set>"\s*,/.test(allowedCell),
+      `agent_skills Allowed-Values cell must show the array-of-strings form, got: ${allowedCell}`
+    );
+    assert.ok(
+      /array/i.test(cells[5]),
+      'agent_skills description must explain the array form assigns multiple skill sets to one agent type'
+    );
+  });
+});
+
 // ─── CONFIGURATION.md parity (#1216) ────────────────────────────────────────
 
 describe('CONFIGURATION.md parity (#1216)', () => {