Phase 1 Workstream A+E: thesis substrate + dual-approval gate

- migration 0002_phase1_architect: thesis_lines (core spine + per-segment lines),
  thesis_nodes (+ append-only revisions), thesis_versions (one-canonical-per-line
  DB invariant), thesis_reviews (dual approval + feedback), segments. Reversible.
- backend/mcp/architect_tools.py: agent draft tools (node tree, versions,
  segments, get_canonical fails-closed) — NO self-approval path. MCP-exposed.
- backend/thesis_review.py + server.py routes: human-gated approval. Dual sign-off
  via thesis_required_approvals; atomic supersede; every action logged.
- docs/PHASE_1.md (kickoff brief); docs/OPERATIONS.md (partner guide);
  start9/0.4 "Resolve duplicate names" fuzzy action.

Verified on synthetic data: dual approval promotes correctly, exactly one
canonical survives supersede, get_canonical fails closed, full interaction_log.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Keysat
2026-06-05 10:20:00 -05:00
parent 6be2e40f54
commit 3e199fd8d5
10 changed files with 993 additions and 0 deletions
+281
View File
@@ -0,0 +1,281 @@
"""Architect MCP tool logic (Phase 1, Workstream A/E) — plain, testable functions.
The Architect drafts and iterates on the thesis; it CANNOT make anything canonical
— promotion to canonical is a human-only action on a CRM HTTP route (server.py),
not exposed here (guardrail #4). Every write goes through interaction_log
(guardrail #5). Mirrors crm_tools.py conventions.
Tool surface:
reads list_thesis_lines, get_thesis, get_node, get_node_history,
list_versions, get_canonical_thesis, get_review_feedback,
list_segments, get_segment
drafts create_thesis_line, upsert_thesis_node, create_thesis_version,
submit_version_for_review, upsert_segment
NO approve/promote/publish/outbound tool exists.
"""
import json
import os
import sqlite3
import sys
import uuid
from datetime import datetime, timezone
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "ingest"))
import config # noqa: E402
def _conn(db=None):
c = sqlite3.connect(db or os.environ.get("CRM_DB_PATH") or config.DEFAULT_DB)
c.row_factory = sqlite3.Row
c.execute("PRAGMA foreign_keys=ON")
return c
def _now():
return datetime.now(timezone.utc).replace(tzinfo=None).isoformat() + "Z"
def _eid(prefix):
return f"{prefix}_{uuid.uuid4().hex[:16]}"
def _log(c, action, target_id, payload, actor_id="architect", actor_type="agent"):
c.execute("""INSERT INTO interaction_log
(id, ts, actor_type, actor_id, action, target_type, target_id, payload, source, created_at)
VALUES (?,?,?,?,?,?,?,?,?,?)""",
(str(uuid.uuid4()), _now(), actor_type, actor_id, action, "thesis", target_id,
json.dumps(payload) if payload is not None else None, "architect", _now()))
def _line_by_key(c, line_key):
return c.execute("SELECT * FROM thesis_lines WHERE line_key=? AND deleted_at IS NULL", (line_key,)).fetchone()
# ── reads ─────────────────────────────────────────────────────────────────────
def list_thesis_lines(db=None):
c = _conn(db)
rows = [dict(r) for r in c.execute(
"SELECT id, line_key, name, segment_key, is_core, status FROM thesis_lines WHERE deleted_at IS NULL ORDER BY is_core DESC, name")]
c.close()
return {"lines": rows, "count": len(rows)}
def _node_tree(c, line_id):
nodes = [dict(r) for r in c.execute(
"SELECT * FROM thesis_nodes WHERE line_id=? AND deleted_at IS NULL ORDER BY ord", (line_id,))]
by_parent = {}
for n in nodes:
by_parent.setdefault(n["parent_id"], []).append(n)
def build(pid):
out = []
for n in by_parent.get(pid, []):
out.append({**{k: n[k] for k in ("id", "node_type", "title", "body", "status", "variant_group", "ord")},
"children": build(n["id"])})
return out
return build(None)
def get_thesis(line_key, db=None):
"""A thesis line + its node tree."""
c = _conn(db)
line = _line_by_key(c, line_key)
if not line:
c.close()
return {"error": "not_found", "line_key": line_key}
out = {"line": dict(line), "tree": _node_tree(c, line["id"])}
c.close()
return out
def get_node(node_id, db=None):
c = _conn(db)
r = c.execute("SELECT * FROM thesis_nodes WHERE id=?", (node_id,)).fetchone()
c.close()
return dict(r) if r else {"error": "not_found", "node_id": node_id}
def get_node_history(node_id, db=None):
c = _conn(db)
rows = [dict(r) for r in c.execute(
"SELECT rev_no, body, title, status, change_summary, change_reason, actor_type, actor_id, created_at "
"FROM thesis_node_revisions WHERE node_id=? ORDER BY rev_no DESC", (node_id,))]
c.close()
return {"node_id": node_id, "revisions": rows}
def list_versions(line_key, db=None):
c = _conn(db)
line = _line_by_key(c, line_key)
if not line:
c.close()
return {"error": "not_found", "line_key": line_key}
rows = [dict(r) for r in c.execute(
"SELECT id, version_no, status, rationale, created_by, created_at, approved_at "
"FROM thesis_versions WHERE line_id=? ORDER BY version_no DESC", (line["id"],))]
c.close()
return {"line_key": line_key, "versions": rows}
def get_canonical_thesis(line_key, db=None):
"""The single canonical version's body_json. FAILS CLOSED if none approved —
so Scribe/downstream agents can never generate against an unapproved thesis."""
c = _conn(db)
line = _line_by_key(c, line_key)
if not line:
c.close()
return {"status": "no_such_line", "line_key": line_key}
r = c.execute("SELECT * FROM thesis_versions WHERE line_id=? AND status='canonical'", (line["id"],)).fetchone()
c.close()
if not r:
return {"status": "no_canonical_thesis", "line_key": line_key}
return {"status": "ok", "line_key": line_key, "version_id": r["id"], "version_no": r["version_no"],
"approved_at": r["approved_at"], "thesis": json.loads(r["body_json"])}
def get_review_feedback(version_id, db=None):
"""Partners' reviews/feedback on a version — what the Architect iterates on."""
c = _conn(db)
rows = [dict(r) for r in c.execute(
"SELECT reviewer_user_id, decision, feedback, target_node_id, created_at "
"FROM thesis_reviews WHERE version_id=? ORDER BY created_at", (version_id,))]
approvals = sum(1 for r in rows if r["decision"] == "approve")
c.close()
return {"version_id": version_id, "reviews": rows, "approvals": approvals}
def list_segments(db=None):
c = _conn(db)
rows = [dict(r) for r in c.execute(
"SELECT segment_key, name, definition, needs_to_hear, avoid, version_no FROM segments WHERE status='active' ORDER BY name")]
c.close()
return {"segments": rows, "count": len(rows)}
def get_segment(segment_key, db=None):
c = _conn(db)
r = c.execute("SELECT * FROM segments WHERE segment_key=? AND status='active'", (segment_key,)).fetchone()
c.close()
return dict(r) if r else {"error": "not_found", "segment_key": segment_key}
# ── draft writes (logged; never canonical) ────────────────────────────────────
def create_thesis_line(line_key, name, segment_key=None, is_core=False, description=None, db=None):
c = _conn(db)
lid = _eid("thl")
c.execute("""INSERT INTO thesis_lines (id, line_key, name, segment_key, is_core, description, created_at, updated_at)
VALUES (?,?,?,?,?,?,?,?)""",
(lid, line_key, name, segment_key, 1 if is_core else 0, description, _now(), _now()))
_log(c, "thesis.line_created", lid, {"line_key": line_key, "segment_key": segment_key, "is_core": bool(is_core)})
c.commit()
c.close()
return {"id": lid, "line_key": line_key}
def upsert_thesis_node(line_id, node_type, body, title=None, parent_id=None, ord=None,
variant_group=None, node_id=None, change_reason=None, change_summary=None,
actor_id="architect", claude_session_id=None, meta=None, db=None):
"""Create or edit a node. On edit, the prior state is written to
thesis_node_revisions before the live row changes (full provenance)."""
c = _conn(db)
if node_id:
prev = c.execute("SELECT * FROM thesis_nodes WHERE id=?", (node_id,)).fetchone()
if not prev:
c.close()
return {"error": "not_found", "node_id": node_id}
rev_no = (c.execute("SELECT COALESCE(MAX(rev_no),0) FROM thesis_node_revisions WHERE node_id=?",
(node_id,)).fetchone()[0]) + 1
c.execute("""INSERT INTO thesis_node_revisions
(id, node_id, line_id, rev_no, node_type, title, body, status, ord, variant_group, meta,
change_summary, change_reason, actor_type, actor_id, claude_session_id, created_at)
VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)""",
(str(uuid.uuid4()), node_id, prev["line_id"], rev_no, prev["node_type"], prev["title"], prev["body"],
prev["status"], prev["ord"], prev["variant_group"], prev["meta"], change_summary, change_reason,
"agent", actor_id, claude_session_id, _now()))
c.execute("""UPDATE thesis_nodes SET node_type=?, title=COALESCE(?,title), body=?, ord=COALESCE(?,ord),
variant_group=?, meta=COALESCE(?,meta), updated_at=? WHERE id=?""",
(node_type, title, body, ord, variant_group, json.dumps(meta) if meta else None, _now(), node_id))
_log(c, "thesis.node_revised", node_id, {"line_id": prev["line_id"], "rev_no": rev_no, "reason": change_reason})
out = {"id": node_id, "rev_no": rev_no}
else:
nid = _eid("thn")
if ord is None:
ord = (c.execute("SELECT COALESCE(MAX(ord),0) FROM thesis_nodes WHERE line_id=? AND parent_id IS ?",
(line_id, parent_id)).fetchone()[0]) + 1.0
c.execute("""INSERT INTO thesis_nodes (id, line_id, parent_id, node_type, ord, title, body, status, variant_group, meta, created_at, updated_at)
VALUES (?,?,?,?,?,?,?, 'draft', ?,?,?,?)""",
(nid, line_id, parent_id, node_type, ord, title, body, variant_group,
json.dumps(meta) if meta else None, _now(), _now()))
_log(c, "thesis.node_created", nid, {"line_id": line_id, "node_type": node_type})
out = {"id": nid, "rev_no": 0}
c.commit()
c.close()
return out
def create_thesis_version(line_key, rationale=None, created_by="architect", db=None):
"""Freeze the current node tree of a line into an immutable draft version
(body_json = the Architect->Scribe contract). Stays 'draft' until submitted
and human-approved."""
c = _conn(db)
line = _line_by_key(c, line_key)
if not line:
c.close()
return {"error": "not_found", "line_key": line_key}
tree = _node_tree(c, line["id"])
# typed projection for the Scribe contract
flat = [dict(r) for r in c.execute(
"SELECT node_type, title, body FROM thesis_nodes WHERE line_id=? AND deleted_at IS NULL ORDER BY ord", (line["id"],))]
def of(t):
return [{"title": n["title"], "body": n["body"]} for n in flat if n["node_type"] == t]
body_json = {
"line_key": line_key, "name": line["name"], "segment_key": line["segment_key"],
"throughline": of("throughline"), "pillars": of("section"), "claims": of("claim"),
"proof_points": of("proof_point"), "objections": of("objection"), "segment_cuts": of("segment_cut"),
"tree": tree, "generated_at": _now(),
}
vno = (c.execute("SELECT COALESCE(MAX(version_no),0) FROM thesis_versions WHERE line_id=?",
(line["id"],)).fetchone()[0]) + 1
vid = _eid("thv")
c.execute("""INSERT INTO thesis_versions (id, line_id, version_no, body_json, status, rationale, created_by, created_at)
VALUES (?,?,?,?, 'draft', ?,?,?)""",
(vid, line["id"], vno, json.dumps(body_json), rationale, created_by, _now()))
_log(c, "thesis.version_created", vid, {"line_key": line_key, "version_no": vno})
c.commit()
c.close()
return {"id": vid, "version_no": vno, "status": "draft"}
def submit_version_for_review(version_id, db=None):
c = _conn(db)
r = c.execute("SELECT status FROM thesis_versions WHERE id=?", (version_id,)).fetchone()
if not r:
c.close()
return {"error": "not_found", "version_id": version_id}
if r["status"] != "draft":
c.close()
return {"error": "not_draft", "status": r["status"]}
c.execute("UPDATE thesis_versions SET status='in_review' WHERE id=?", (version_id,))
_log(c, "thesis.submitted_for_review", version_id, None)
c.commit()
c.close()
return {"version_id": version_id, "status": "in_review"}
def upsert_segment(segment_key, name, definition=None, needs_to_hear=None, avoid=None, db=None):
"""Create/replace a segment's active definition (retire the prior active row)."""
c = _conn(db)
prev = c.execute("SELECT version_no FROM segments WHERE segment_key=? AND status='active'", (segment_key,)).fetchone()
vno = (prev["version_no"] + 1) if prev else 1
if prev:
c.execute("UPDATE segments SET status='retired', updated_at=? WHERE segment_key=? AND status='active'",
(_now(), segment_key))
sid = _eid("seg")
c.execute("""INSERT INTO segments (id, segment_key, name, definition, needs_to_hear, avoid, version_no, status, created_at, updated_at)
VALUES (?,?,?,?,?,?,?, 'active', ?,?)""",
(sid, segment_key, name, definition, needs_to_hear, avoid, vno, _now(), _now()))
_log(c, "segment.upserted", sid, {"segment_key": segment_key, "version_no": vno})
c.commit()
c.close()
return {"id": sid, "segment_key": segment_key, "version_no": vno}