diff --git a/dist/.devtrail/audit-prompts/auditor-primary.md b/dist/.devtrail/audit-prompts/auditor-primary.md
new file mode 100644
index 0000000..e873c00
--- /dev/null
+++ b/dist/.devtrail/audit-prompts/auditor-primary.md
@@ -0,0 +1,154 @@
+<!--
+DevTrail audit prompt — auditor-primary role.
+
+This file is a TEMPLATE. `devtrail charter audit <CHARTER-ID>` resolves the
+placeholders below against the Charter's content + git range + originating
+AILOGs, and writes the resolved prompt to:
+
+    audit/charters/<CHARTER-ID>/prompts/auditor-primary.prompt.md
+
+The resolved prompt is what the operator pastes into their auditor of choice
+(e.g., a Copilot, Gemini, or Claude chat). The auditor's response is saved
+to:
+
+    audit/charters/<CHARTER-ID>/auditor-primary.md
+
+Adopters may edit this template to suit their project's conventions; the CLI
+will use whatever lives at `.devtrail/audit-prompts/auditor-primary.md` at
+prompt-resolution time. Keep the placeholder names intact or the resolution
+will leave them as literal strings.
+
+Placeholders supported by `devtrail charter audit`:
+  {{charter_id}}        — e.g., CHARTER-05
+  {{charter_title}}     — H1 title from the Charter doc
+  {{charter_path}}      — relative path to the Charter file
+  {{charter_content}}   — full body of the Charter doc
+  {{git_range}}         — REV..REV that bounds the audit
+  {{git_diff}}          — output of `git diff <git_range>`
+  {{ailog_paths}}       — newline-separated list of originating_ailogs paths
+  {{ailog_contents}}    — concatenated bodies of those AILOGs
+  {{audit_role}}        — for this template, always "auditor-primary"
+  {{schema_path}}       — relative path to audit-output.schema.v0.json
+-->
+
+You are an external auditor reviewing the execution of a DevTrail Charter.
+Your job is to compare what the Charter declared (ex-ante) against what the
+commits actually changed (ex-post) and produce a categorized list of findings.
+
+You are the **{{audit_role}}** auditor in a dual-audit cycle. Another
+auditor of a different model family is being given the same Charter and diff
+in parallel. A calibrator-reconciler will later compare your findings against
+theirs. Cross-model heterogeneity is the point — your distribution of
+training and your blind spots differ from the other auditor's, and that is
+what makes the convergence (or disagreement) signal valuable.
+
+# What you are auditing
+
+**Charter:** `{{charter_path}}` (`{{charter_id}}` — {{charter_title}})
+
+**Git range:** `{{git_range}}`
+
+**Originating AILOGs** (rationale + emergent risks documented during execution):
+
+```
+{{ailog_paths}}
+```
+
+# Charter content
+
+```markdown
+{{charter_content}}
+```
+
+# AILOG content
+
+```markdown
+{{ailog_contents}}
+```
+
+# Diff
+
+```diff
+{{git_diff}}
+```
+
+# What I need from you
+
+Produce a markdown file with this exact frontmatter shape (validates against
+`{{schema_path}}`):
+
+```yaml
+---
+audit_role: auditor-primary
+auditor: <your model id and version>      # e.g., copilot-v1.0.37
+charter_id: {{charter_id}}
+git_range: "{{git_range}}"
+prompt_used: prompts/auditor-primary.prompt.md
+audited_at: <today YYYY-MM-DD>
+findings_total: <count>
+findings_by_category:
+  hallucination: <count>
+  implementation_gap: <count>
+  real_debt: <count>
+  false_positive: <count>
+---
+
+# Audit: {{charter_id}} by <your model id>
+
+## Summary
+
+[1-2 paragraphs: did the execution match the Charter's declared scope? What
+is the overall verdict — clean, partial, deviated?]
+
+## Findings
+
+### F1 — <short title> — <category>
+
+**Where:** `<file:line>` or `<file>` if span-wide.
+
+**What I observed:** [Concrete description of the gap, hallucination, or
+real debt. Cite specific lines from the diff or the AILOGs.]
+
+**Why I'm flagging it:** [Reasoning. What about the Charter's declaration vs
+the diff makes this a finding?]
+
+### F2 — ...
+
+[Continue numbering F1...FN. One section per finding.]
+```
+
+# Categorization rules
+
+Apply the following categories. The calibrator will use the same definitions:
+
+- **`hallucination`** — the Charter or implementation references something
+  that does not exist (an API, a function, a field name, a behavior). The
+  agent invented it. Verify by reading the diff or the cited file.
+- **`implementation_gap`** — the Charter declared work that the diff did
+  not deliver, OR the diff delivered work the Charter did not declare,
+  WITHOUT it being documented as drift in the AILOG. (If documented in
+  AILOG under `## Risk` as `R<N+1>`, that is *not* a gap; the AILOG-aware
+  drift check already accepts it.)
+- **`real_debt`** — code-level concern that is correct as far as the
+  Charter goes but introduces technical debt or a subtle defect (a missing
+  error path, a leaky resource, a non-idempotent operation). Adopter is
+  expected to capture as `TDE` doc post-audit.
+- **`false_positive`** — what initially looked like a finding but, on
+  closer inspection of the AILOGs or the diff context, isn't one.
+  Document anyway; the calibrator uses these to recognize patterns where
+  one auditor over-reports.
+
+# Discipline
+
+- Cite specific file paths and line numbers from the diff. Do not summarize
+  abstractly.
+- If you cannot find anything substantive, return `findings_total: 0` with
+  a single `## Summary` paragraph explaining what you reviewed. Empty audits
+  are valid signal — the calibrator will note convergence with the other
+  auditor's empty audit, if applicable.
+- Do not fabricate findings to seem thorough. The categorization rules
+  above include `false_positive` precisely because over-reporting is a
+  real audit failure mode.
+- Do not consult external sources beyond what is provided in this prompt.
+  The audit must be reproducible from the prompt + the diff + the AILOGs
+  alone.
diff --git a/dist/.devtrail/audit-prompts/auditor-secondary.md b/dist/.devtrail/audit-prompts/auditor-secondary.md
new file mode 100644
index 0000000..a2651fd
--- /dev/null
+++ b/dist/.devtrail/audit-prompts/auditor-secondary.md
@@ -0,0 +1,131 @@
+<!--
+DevTrail audit prompt — auditor-secondary role.
+
+Mirror of auditor-primary.md with `audit_role: auditor-secondary` and a
+deliberately different framing in the introduction. The body of the prompt
+is intentionally structurally identical so that the calibrator-reconciler
+can compare findings symmetrically — the "heterogeneity" signal lives in
+the auditor MODEL (different model family per §5.2), not in different
+prompts.
+
+If you ever need to A/B-test prompt phrasings between primary and
+secondary, do it deliberately and document the asymmetry here.
+
+Placeholders are the same set as auditor-primary.md. See that file's header
+for the full list.
+-->
+
+You are an independent external auditor reviewing the execution of a
+DevTrail Charter. You are the **{{audit_role}}** auditor. A primary auditor
+of a different model family is reviewing the same Charter and diff in
+parallel. The two of you may agree or disagree; both are valuable signal.
+A calibrator-reconciler will integrate your findings with the primary's.
+
+You may have been trained on different data than the primary. Your blind
+spots and your priors are different. Audit independently — the value of the
+dual-audit comes from convergence on real findings and divergence on
+boundary cases, not from echoing the primary auditor.
+
+# What you are auditing
+
+**Charter:** `{{charter_path}}` (`{{charter_id}}` — {{charter_title}})
+
+**Git range:** `{{git_range}}`
+
+**Originating AILOGs** (rationale + emergent risks documented during execution):
+
+```
+{{ailog_paths}}
+```
+
+# Charter content
+
+```markdown
+{{charter_content}}
+```
+
+# AILOG content
+
+```markdown
+{{ailog_contents}}
+```
+
+# Diff
+
+```diff
+{{git_diff}}
+```
+
+# What I need from you
+
+Produce a markdown file with this exact frontmatter shape (validates against
+`{{schema_path}}`):
+
+```yaml
+---
+audit_role: auditor-secondary
+auditor: <your model id and version>      # e.g., gemini-cli-v1.5
+charter_id: {{charter_id}}
+git_range: "{{git_range}}"
+prompt_used: prompts/auditor-secondary.prompt.md
+audited_at: <today YYYY-MM-DD>
+findings_total: <count>
+findings_by_category:
+  hallucination: <count>
+  implementation_gap: <count>
+  real_debt: <count>
+  false_positive: <count>
+---
+
+# Audit: {{charter_id}} by <your model id>
+
+## Summary
+
+[1-2 paragraphs: did the execution match the Charter's declared scope?
+What is the overall verdict?]
+
+## Findings
+
+### F1 — <short title> — <category>
+
+**Where:** `<file:line>` or `<file>` if span-wide.
+
+**What I observed:** [Concrete description. Cite specific lines from the
+diff or the AILOGs.]
+
+**Why I'm flagging it:** [Reasoning. What about the Charter's declaration
+vs the diff makes this a finding?]
+
+### F2 — ...
+
+[One section per finding.]
+```
+
+# Categorization rules
+
+Same categories as the primary auditor — the calibrator uses the same
+definitions to compare your findings:
+
+- **`hallucination`** — Charter or implementation references something
+  that does not exist (invented API, function, field, behavior). Verify
+  by reading the diff or cited file.
+- **`implementation_gap`** — Charter declared work the diff did not
+  deliver (or vice versa) WITHOUT it being documented as drift in the
+  AILOG. (Documented in AILOG `## Risk` as `R<N+1>` is *not* a gap.)
+- **`real_debt`** — code-level concern not strictly within Charter
+  scope but introducing debt or a subtle defect (missing error path,
+  leaky resource, non-idempotent operation). Adopter captures as `TDE`.
+- **`false_positive`** — looked like a finding but, on closer reading
+  of the AILOGs or diff context, isn't. Document anyway; calibrator
+  uses these to detect over-reporting patterns.
+
+# Discipline
+
+- Cite specific file paths and line numbers from the diff. No abstract
+  summaries.
+- If you find nothing substantive, return `findings_total: 0` with a
+  `## Summary` paragraph explaining your review. Empty is valid signal.
+- Do not fabricate findings to seem thorough. Over-reporting is a real
+  audit failure mode — `false_positive` exists precisely for this case.
+- Do not consult external sources beyond this prompt. The audit must be
+  reproducible from the prompt + diff + AILOGs alone.
diff --git a/dist/.devtrail/audit-prompts/calibrator-reconciler.md b/dist/.devtrail/audit-prompts/calibrator-reconciler.md
new file mode 100644
index 0000000..1dd4763
--- /dev/null
+++ b/dist/.devtrail/audit-prompts/calibrator-reconciler.md
@@ -0,0 +1,173 @@
+<!--
+DevTrail audit prompt — calibrator-reconciler role.
+
+Phase 3 §5.1 of the roadmap describes the three-tier audit: two heterogeneous
+auditors in parallel + one calibrator-reconciler that reads BOTH outputs and
+applies the categorization schema definitionally. The calibrator's job is
+not to discover gaps (the auditors did that); it is to:
+  1. Recognize agreement (both auditors flagged the same thing → high-signal).
+  2. Resolve disagreement (one flagged, the other didn't, or they disagree
+     on category → calibrator decides per the schema rules).
+  3. Reject false positives (both auditors flagged but the AILOGs document
+     it, or diff context makes it not-a-finding).
+
+Per §5.2, the calibrator MAY be of the same family as the implementer
+because its task is definitional, not discovery. Heterogeneity matters for
+the auditor pair only.
+
+This file is a TEMPLATE. `devtrail charter audit <CHARTER-ID> --calibrate`
+resolves the placeholders below and writes the resolved prompt to:
+
+    audit/charters/<CHARTER-ID>/prompts/calibrator-reconciler.prompt.md
+
+The operator runs the resolved prompt in their calibrator of choice and saves
+the response to:
+
+    audit/charters/<CHARTER-ID>/calibrator-reconciler.md
+
+Placeholders supported (in addition to the standard set listed in
+auditor-primary.md):
+  {{auditor_primary_findings}}    — full content of auditor-primary.md
+  {{auditor_secondary_findings}}  — full content of auditor-secondary.md
+  {{auditors_reconciled}}         — array shape for the frontmatter
+-->
+
+You are the **calibrator-reconciler** of a DevTrail dual-audit cycle. Two
+external auditors of different model families have already reviewed the
+Charter; their outputs are below. Your job is to apply the categorization
+schema definitionally, recognize agreement and disagreement, and produce a
+consolidated list of findings that the Charter's telemetry can record.
+
+You are not auditing fresh. You are reading two audits and reconciling them.
+
+# What you are reconciling
+
+**Charter:** `{{charter_path}}` (`{{charter_id}}` — {{charter_title}})
+
+**Git range:** `{{git_range}}`
+
+# Charter content
+
+```markdown
+{{charter_content}}
+```
+
+# Originating AILOGs
+
+```
+{{ailog_paths}}
+```
+
+```markdown
+{{ailog_contents}}
+```
+
+# Auditor PRIMARY output
+
+```markdown
+{{auditor_primary_findings}}
+```
+
+# Auditor SECONDARY output
+
+```markdown
+{{auditor_secondary_findings}}
+```
+
+# What I need from you
+
+Produce a markdown file with this exact frontmatter shape (validates against
+`{{schema_path}}`):
+
+```yaml
+---
+audit_role: calibrator-reconciler
+calibrator: <your model id and version>      # e.g., claude-opus-4
+charter_id: {{charter_id}}
+git_range: "{{git_range}}"
+prompt_used: prompts/calibrator-reconciler.prompt.md
+calibrated_at: <today YYYY-MM-DD>
+auditors_reconciled:
+  - auditor-primary.md
+  - auditor-secondary.md
+findings_consolidated: <count>
+findings_by_status:
+  agreed: <count>            # both auditors flagged the same finding
+  disputed: <count>          # both flagged but disagreed on category — you picked
+  unique_primary: <count>    # only primary; you validated as legitimate
+  unique_secondary: <count>  # only secondary; you validated
+  rejected: <count>          # both flagged but you determined false positive
+---
+
+# Calibration: {{charter_id}}
+
+## Reconciliation summary
+
+[1-2 paragraphs: how convergent were the auditors? Where did they
+disagree, and on what kind of finding? Did one auditor have a higher
+false-positive rate?]
+
+## Reconciled findings
+
+### C1 — <short title> — <category> — <status>
+
+**Status:** agreed | disputed | unique_primary | unique_secondary | rejected.
+
+**Where:** `<file:line>`.
+
+**What was observed:** [Combine the auditors' descriptions. If they
+disagreed, note both views and your resolution.]
+
+**Calibration rationale:** [Why this status. If `agreed`, name what each
+auditor said. If `disputed`, name the disagreement and your call. If
+`unique_*`, explain why you validated. If `rejected`, explain why both
+auditors were wrong.]
+
+### C2 — ...
+
+[One section per consolidated finding. Numbering C1...CN is independent
+of the F1...FN numbering each auditor used; cross-reference auditor
+numbering inside each section as needed.]
+```
+
+# Categorization rules (same as the auditors)
+
+- **`hallucination`** — invented API, function, field, behavior.
+- **`implementation_gap`** — declared but not delivered (or vice versa)
+  WITHOUT being documented in AILOG as drift.
+- **`real_debt`** — code-level debt or subtle defect outside Charter scope.
+- **`false_positive`** — appeared to be a finding but isn't.
+
+# Status assignment rules
+
+For each distinct finding (deduplicate when both auditors describe the
+same gap with different wording):
+
+- `agreed` — both auditors flagged it AND assigned the same category.
+  Strongest signal — the convergence between heterogeneous auditors is
+  what makes a dual-audit valuable.
+- `disputed` — both auditors flagged it BUT assigned different categories
+  (e.g., primary calls it `implementation_gap`, secondary calls it
+  `hallucination`). You pick the category that fits the schema definitions
+  best, given the diff and the AILOGs.
+- `unique_primary` / `unique_secondary` — only one auditor flagged it,
+  AND on your reading, they were correct to flag it.
+- `rejected` — one or both auditors flagged it, but on closer reading
+  of the AILOGs (especially `## Risk` `R<N+1>` documented mitigations)
+  or the diff, it isn't a finding. Both `unique` flags can become
+  `rejected` if the unique auditor was wrong.
+
+# Discipline
+
+- Use the `findings_by_status` counts as a cross-check against your
+  body sections. They must add up to `findings_consolidated`.
+- Do not introduce findings the auditors did not see. If you spot
+  something they missed, document it in `## Reconciliation summary` as
+  an observation, not as a `C<N>` finding. Fresh findings are out of
+  scope for the calibrator role — that's what the next audit cycle is for.
+- The `rejected` count is signal worth tracking — it tells the Charter
+  author which audit categories tend to over-report on this kind of
+  Charter, which improves future audit prompt design.
+- Do not consult external sources beyond what is provided. The
+  reconciliation must be reproducible from the prompt + the two auditor
+  outputs + the Charter + the AILOGs.
diff --git a/dist/.devtrail/schemas/audit-output.schema.v0.json b/dist/.devtrail/schemas/audit-output.schema.v0.json
new file mode 100644
index 0000000..b629138
--- /dev/null
+++ b/dist/.devtrail/schemas/audit-output.schema.v0.json
@@ -0,0 +1,164 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://devtrail.dev/schemas/audit-output.schema.v0.json",
+  "title": "DevTrail Charter Audit Output (experimental v0)",
+  "description": "Frontmatter schema for the markdown files that auditors and the calibrator-reconciler produce during a `devtrail charter audit` cycle. The shape distinguishes auditor outputs (primary/secondary, fresh findings) from calibrator outputs (reconciliation across auditor outputs) via the `audit_role` field — one of three fixed roles, not arbitrary N.",
+  "$comment": "EXPERIMENTAL v0. Phase 3 of the CLI roadmap. Crystallized from the dual-audit pattern validated empirically in Sentinel (Copilot + Gemini + claude-analisis across 6 cycles). Will not stabilize to v1.0 until validated in a second domain. The schema is for the YAML inside the file's frontmatter; the body is free-form markdown by convention. The findings_by_category enum (hallucination | implementation_gap | real_debt | false_positive) is the same one used by `external_audit` in charter-telemetry.schema.v0.json — the audit cycle output integrates directly into the Charter telemetry at close time.",
+  "type": "object",
+  "oneOf": [
+    { "$ref": "#/$defs/auditorOutput" },
+    { "$ref": "#/$defs/calibratorOutput" }
+  ],
+  "$defs": {
+    "auditorOutput": {
+      "type": "object",
+      "required": [
+        "audit_role",
+        "auditor",
+        "charter_id",
+        "git_range",
+        "prompt_used",
+        "audited_at",
+        "findings_total",
+        "findings_by_category"
+      ],
+      "additionalProperties": true,
+      "properties": {
+        "audit_role": {
+          "type": "string",
+          "enum": ["auditor-primary", "auditor-secondary"],
+          "description": "Slot this output occupies in the dual-audit. Primary and secondary are heterogeneous by design (see CLI-REFERENCE.md `devtrail charter audit` for the inter-family recommendation)."
+        },
+        "auditor": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Model identifier and version of the auditor that produced this output (e.g., `copilot-v1.0.37`, `gemini-cli-v1.5`, `claude-opus-4`). Free-form so adopters can use any model identifier."
+        },
+        "charter_id": {
+          "type": "string",
+          "pattern": "^CHARTER-[0-9]{2,}(-[a-z0-9-]+)?$"
+        },
+        "git_range": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Git revision range the auditor reviewed (e.g., `origin/main..HEAD`)."
+        },
+        "prompt_used": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Path (relative to the audit directory) of the resolved prompt file the auditor was given. Required so the operator can verify the prompt wasn't tampered with mid-cycle. Closes RFC #82."
+        },
+        "audited_at": {
+          "type": "string",
+          "format": "date"
+        },
+        "findings_total": {
+          "type": "integer",
+          "minimum": 0
+        },
+        "findings_by_category": {
+          "type": "object",
+          "additionalProperties": false,
+          "properties": {
+            "hallucination": { "type": "integer", "minimum": 0 },
+            "implementation_gap": { "type": "integer", "minimum": 0 },
+            "real_debt": { "type": "integer", "minimum": 0 },
+            "false_positive": { "type": "integer", "minimum": 0 }
+          }
+        },
+        "audit_quality": {
+          "type": "string",
+          "enum": ["high", "medium", "low"],
+          "description": "Operator's calibration of this auditor's output quality. Optional — set when consolidating into telemetry."
+        }
+      }
+    },
+    "calibratorOutput": {
+      "type": "object",
+      "required": [
+        "audit_role",
+        "calibrator",
+        "charter_id",
+        "git_range",
+        "prompt_used",
+        "calibrated_at",
+        "auditors_reconciled",
+        "findings_consolidated"
+      ],
+      "additionalProperties": true,
+      "properties": {
+        "audit_role": {
+          "type": "string",
+          "const": "calibrator-reconciler"
+        },
+        "calibrator": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Model identifier of the calibrator. May be of any family — even the same as the implementer — because the calibrator's job is to apply the schema definitionally over already-produced auditor verdicts, not to discover gaps. Heterogeneity matters for the auditor pair, not the calibrator (see roadmap §5.2)."
+        },
+        "charter_id": {
+          "type": "string",
+          "pattern": "^CHARTER-[0-9]{2,}(-[a-z0-9-]+)?$"
+        },
+        "git_range": {
+          "type": "string",
+          "minLength": 1
+        },
+        "prompt_used": {
+          "type": "string",
+          "minLength": 1
+        },
+        "calibrated_at": {
+          "type": "string",
+          "format": "date"
+        },
+        "auditors_reconciled": {
+          "type": "array",
+          "minItems": 2,
+          "maxItems": 2,
+          "items": {
+            "type": "string",
+            "minLength": 1,
+            "description": "Path (relative to audit dir) of an auditor output file this calibration considered."
+          },
+          "description": "Exactly two auditor outputs are reconciled in v0 (dual-audit pattern). N≥3 audit ensembles are forward-looking."
+        },
+        "findings_consolidated": {
+          "type": "integer",
+          "minimum": 0,
+          "description": "Total distinct findings after reconciliation (after deduplication and rejection of false positives identified across both auditors)."
+        },
+        "findings_by_status": {
+          "type": "object",
+          "additionalProperties": false,
+          "properties": {
+            "agreed": {
+              "type": "integer",
+              "minimum": 0,
+              "description": "Both auditors flagged the same finding (highest signal — convergence)."
+            },
+            "disputed": {
+              "type": "integer",
+              "minimum": 0,
+              "description": "Auditors flagged but disagreed on category; calibrator picked one."
+            },
+            "unique_primary": {
+              "type": "integer",
+              "minimum": 0,
+              "description": "Only the primary auditor flagged this; calibrator validated as legitimate."
+            },
+            "unique_secondary": {
+              "type": "integer",
+              "minimum": 0
+            },
+            "rejected": {
+              "type": "integer",
+              "minimum": 0,
+              "description": "Both auditors flagged but calibrator determined false positive."
+            }
+          }
+        }
+      }
+    }
+  }
+}