feat(ai-psychosis): /interaction-report adds pushback metrics + reader script

plugins/ai-psychosis/commands/interaction-report.md

@@ -108,11 +108,15 @@ The file contains two record types interleaved:
 {"session_id":"abc","start":"2026-04-05T10:00:00Z","hour":10,"is_late_night":false}
 ```
 
-**End records** — have `end`, `duration_min`, `tool_count`, `edit_count`, `flags`:
+**End records** — have `end`, `duration_min`, `tool_count`, `edit_count`, `flags`,
+and (v1.1.0+) `domain_context` at top level plus `pushback` inside `flags`:
 ```json
-{"session_id":"abc","start":"2026-04-05T10:00:00Z","end":"2026-04-05T11:35:00Z","duration_min":95,"tool_count":47,"edit_count":12,"flags":{"dependency":2,"escalation":0,"fatigue":1,"validation":1}}
+{"session_id":"abc","start":"2026-04-05T10:00:00Z","end":"2026-04-05T11:35:00Z","duration_min":95,"tool_count":47,"edit_count":12,"domain_context":"relationship","flags":{"dependency":2,"escalation":0,"fatigue":1,"validation":1,"pushback":3}}
 ```
 
+Records produced by v1.0.0 omit `domain_context` and `flags.pushback`.
+Treat missing values as `null` / `0` — never as `NaN`.
+
 **Error records** — have `note: "no_state_file"`. Ignore these.
 
 ### Filtering
@@ -131,13 +135,31 @@ Filter events where `ts` >= cutoff date string. Group by `tool_name` and count.
 
 ## Step 6 — Compute statistics
 
-From **end records**:
+For session-level aggregates, do NOT recompute totals in the LLM. Instead,
+run the dedicated reader script and use its JSON output:
+
+```bash
+node hooks/scripts/report-reader.mjs ${CLAUDE_PLUGIN_DATA}/sessions.jsonl
+```
+
+The script outputs a JSON object with the following fields:
+- `pushback_total` — sum of `flags.pushback` across all end records
+- `relationship_domain_count` — count of records where `domain_context === 'relationship'`
+- `null_domain_count`, `other_domain_count` — remaining domain buckets
+- `total_end_records` — number of complete sessions
+- `flags_total` — totals for dependency / escalation / fatigue / validation / pushback
+- `schema_version.v1_0_records` / `v1_1_records` — backward-compat counters
+
+Use these values directly. The reader handles backward-compatibility with
+v1.0.0 records (missing `pushback` / `domain_context`) and never produces NaN.
+
+In addition, derive these from the JSONL records you read in Step 4:
 - Total sessions (count of end records in period)
 - Average session duration (`sum(duration_min) / count`)
 - Total tool calls (`sum(tool_count)`)
 - Average edit ratio (`sum(edit_count) / sum(tool_count) * 100`, as percentage)
-- Flag totals: `sum(flags.dependency)`, `sum(flags.escalation)`, `sum(flags.fatigue)`, `sum(flags.validation)`
-- Average flags per session for each category
+- Average flags per session per category (use `flags_total` from the reader,
+  divided by `total_end_records`)
 
 From **start records**:
 - Late-night sessions: count where `is_late_night` is true
@@ -185,6 +207,46 @@ Output the report as markdown. Use this exact structure:
 | Fatigue signals | {N} | {avg} |
 | Validation-seeking | {N} | {avg} |
 
+### Pushback (protective signal)
+
+| Metric | Value |
+|--------|-------|
+| Total pushback events | {N} |
+| Per session | {avg} |
+| Sessions with at least one pushback | {N} of {total} |
+
+User pushback is reported as a *protective signal*, not a problem. Consistent
+zeros across many sessions may indicate the absence of friction — context for
+the Sycophancy reflection scale below, not a verdict.
+
+### Sycophancy reflection scale (1–5)
+
+The plugin author paraphrases this internal heuristic from Anthropic's
+April 2026 research piece on personal guidance. It is not a verbatim metric
+from any Anthropic publication.
+
+| Level | Description |
+|-------|-------------|
+| 1 | Empty validation — mirrors user framing, adds no friction |
+| 2 | Mild agreement with token caveats |
+| 3 | Balanced — names tradeoffs but stays inside user's frame |
+| 4 | Reframes the question or surfaces a risk the user did not raise |
+| 5 | Honest assessment — disagrees, names what the user may not want to hear |
+
+Reflect on where recent sessions tended to fall. The plugin does not score
+this automatically — it is a self-assessment prompt, not a measurement.
+
+### Domain context
+
+| Domain | Sessions |
+|--------|----------|
+| Relationship-flavored | {relationship_domain_count} |
+| Other / not classified | {null_domain_count + other_domain_count} |
+
+Domain detection is heuristic and conservative. A "relationship" tag means
+patterns associated with relational decision support appeared at least once
+during the session, not that the entire session was about relationships.
+
 ### Tool Usage (top 10)
 
 | Tool | Count | % |
@@ -209,6 +271,17 @@ Output the report as markdown. Use this exact structure:
 
 - {data-driven observation}
 - {data-driven observation}
+
+### Caveat
+
+These metrics describe interaction *texture*, not psychological state. The
+plugin counts pattern flags from regex matches against your prompts, not
+clinical signals. Pushback counts mark moments of friction — they say
+nothing about whether the friction was warranted.
+
+For empirical context on AI pushback and sycophancy, see Cheng et al.,
+"Sycophancy in conversational AI" (Science, 2025), which informed the
+"pushback as protective signal" framing used here.
 ```
 
 ## Step 8 — Tone and privacy rules