feat(plugin): write-path trigger — offer prior art before code is rewritten
CI & Build / Python lint (push) Successful in 4s
CI & Build / integration (push) Successful in 27s
CI & Build / TypeScript typecheck (push) Successful in 35s
CI & Build / Python tests (push) Successful in 47s
CI & Build / Build & push image (push) Successful in 1m2s

The milestone headline. Auto-inject fires on the operator's prompt; the
moment reuse is actually lost is later, when the agent decides mid-task to
write a helper. A PreToolUse hook on Write|Edit now fires there.

Channel: `additionalContext` with NO permissionDecision, so the note reaches
Claude beside the tool result and the write is never blocked — a recall aid
must not be able to stop the operator's work. Plain stdout would have been
invisible to the model, and deny/ask would have made a nudge into a gate.

Two arms, different in kind:
- BY PLACE — a snippet recorded at this path (or its directory) is prior art
  by definition, not resemblance, so it is neither scored nor thresholded.
  This is what #2083's reverse lookup was built to answer.
- BY MEANING — semantic search restricted to snippets (new `note_type` filter
  on semantic_search_notes) over the code about to be written.
Place ranks first; the top-k cap spans both arms.

Gates carried over from milestone 93 verbatim: threshold, margin, session
dedup, titles-never-bodies. Own `source='write_path'` in retrieval_logs so
precision is tunable separately — the docstring records that the place arm
is unlogged and hands that to #2085.

Its own on/off in Settings but the SAME threshold/top-k: one "how loud may
Scribe be" knob is easier to reason about than two that drift, and splitting
them later is then a data-backed change rather than a guess.

Details worth keeping: the hook sends a REPO-RELATIVE path because that is
how locations are recorded; the git remote resolves to a project and is never
used as the location `repo` filter (different namespaces, would silently
match nothing); the endpoint stays a GET because a read-scoped API key cannot
POST and every other hook depends on that.

plugin.json 0.1.17 -> 0.1.18. Refs #2082, milestone #232.
This commit is contained in:
2026-07-28 09:08:17 -04:00
parent 083944f0fd
commit e0328f2b1c
10 changed files with 810 additions and 3 deletions
+18
View File
@@ -22,6 +22,7 @@ const retentionSaved = ref(false);
const kbInjectEnabled = ref(true);
const kbInjectThreshold = ref("0.55");
const kbInjectTopK = ref("3");
const kbWritePathEnabled = ref(true);
const savingKbInject = ref(false);
const kbInjectSaved = ref(false);
@@ -76,6 +77,9 @@ async function saveKbInject() {
kb_autoinject_enabled: kbInjectEnabled.value ? 'true' : 'false',
kb_autoinject_threshold: String(t),
kb_autoinject_top_k: String(k),
// Its own switch, but deliberately the same threshold/ceiling — see
// WRITEPATH_ENABLED_KEY in services/plugin_context.py.
kb_writepath_enabled: kbWritePathEnabled.value ? 'true' : 'false',
});
kbInjectSaved.value = true;
setTimeout(() => (kbInjectSaved.value = false), 2000);
@@ -459,6 +463,7 @@ onMounted(async () => {
if (allSettings.kb_autoinject_top_k !== undefined) {
kbInjectTopK.value = allSettings.kb_autoinject_top_k;
}
kbWritePathEnabled.value = allSettings.kb_writepath_enabled !== "false";
if (allSettings.notify_task_reminders !== undefined) {
notifyTaskReminders.value = allSettings.notify_task_reminders !== "false";
}
@@ -1193,6 +1198,19 @@ function formatUserDate(iso: string): string {
/>
<p class="field-hint">Ceiling on titles surfaced at once (110).</p>
</div>
<div class="checkbox-field">
<label>
<input type="checkbox" v-model="kbWritePathEnabled" />
Also surface prior art when Claude writes code
</label>
<p class="field-hint">
Checks the file Claude is about to write or edit against your recorded
snippets — what's already kept at that path, and what resembles the code
being written so a helper you already have is offered before it's
rewritten. Uses the same threshold and ceiling above, and never blocks the
edit. Off = prior art surfaces only on your own prompts.
</p>
</div>
<div class="actions">
<button class="btn-save" @click="saveKbInject" :disabled="savingKbInject">
{{ savingKbInject ? 'Saving' : 'Save' }}
+1 -1
View File
@@ -1,7 +1,7 @@
{
"name": "scribe",
"description": "Scribe system-of-record for Claude Code: MCP tools over your notes/tasks/projects/rules, a session-start push channel that surfaces your always-on rules + active-project context, process-skills (writing-plans, systematic-debugging, verification, brainstorming, reusing-code), and your saved Scribe Processes auto-surfaced as skills (/scribe:sync). Replaces superpowers + file-memory with one app-backed plugin.",
"version": "0.1.17",
"version": "0.1.18",
"author": { "name": "Bryan Van Deusen" },
"mcpServers": {
"scribe": {
+12 -2
View File
@@ -7,6 +7,10 @@ instance into a first-class Claude Code extension:
rulebook (the `scribe` server).
- **Session-start push channel** — a `SessionStart` hook injects your always-on
rules + active-project context so Scribe surfaces *without being asked*.
- **Prior-art recall on writes** — a `PreToolUse` hook on Write/Edit checks the
file about to be written against your recorded snippets (what's kept at that
path, and what resembles the code) and offers them before the helper is
rewritten. Titles only, never blocks the edit.
- **Universal process-skills** — using-scribe, writing-plans,
systematic-debugging, verification, brainstorming, reusing-code (record and
recall reusable code as snippets). Replaces superpowers.
@@ -42,6 +46,11 @@ On install you'll be asked for:
- `hooks/hooks.json` → SessionStart hook (`hooks/scribe_session_context.sh`),
**fail-open**: if Scribe is unreachable it injects nothing and never blocks
the session.
- `hooks/hooks.json` → PreToolUse hook on `Write|Edit`
(`hooks/scribe_prior_art.sh`) → `GET /api/plugin/prior-art`. Returns
`additionalContext` with **no** permission decision, so it can inform the write
but never stop it; silent when nothing is recorded, which is most of the time.
Toggle in **Settings → Knowledge auto-inject**.
- `skills/` → the universal process-skills, surfaced by description match.
- `hooks/scribe_sync_processes.sh` (a 2nd SessionStart hook) + the `/scribe:sync`
command → generate `~/.claude/skills/scribe-proc-*` stubs from your Scribe
@@ -52,5 +61,6 @@ On install you'll be asked for:
- Set a `version` bump in `.claude-plugin/plugin.json` per release so clients
pick up changes.
- The session-start hook needs only a **read**-scoped key; the MCP tools need
**write** scope to create/update.
- The session-start, auto-inject and prior-art hooks need only a **read**-scoped
key; the MCP tools need **write** scope to create/update. Every hook is a GET
for that reason — a read key cannot POST.
+11
View File
@@ -23,6 +23,17 @@
}
]
}
],
"PreToolUse": [
{
"matcher": "Write|Edit",
"hooks": [
{
"type": "command",
"command": "bash \"${CLAUDE_PLUGIN_ROOT}/hooks/scribe_prior_art.sh\""
}
]
}
]
}
}
+119
View File
@@ -0,0 +1,119 @@
#!/usr/bin/env bash
# Scribe plugin — PreToolUse write-path trigger (prior-art recall).
#
# Auto-inject (scribe_autoinject.sh) fires on the operator's prompt. The moment
# reuse is actually lost is later: when the AGENT decides mid-task to write a
# helper. This hook fires there — on Write/Edit — and asks the operator's Scribe
# instance what prior art is already recorded for the target file: a snippet at
# that path or in its directory, plus snippets resembling the code about to be
# written. Titles + ids only, never bodies.
#
# NEVER BLOCKS. It returns `additionalContext` with no `permissionDecision`, so
# the write proceeds untouched and Claude sees the note beside the tool result.
# Any failure — unconfigured, unreachable, malformed — exits 0 in silence. A
# recall aid must not be able to stop the operator's work.
#
# Config (same as the other hooks), exported to the hook by Claude Code:
# CLAUDE_PLUGIN_OPTION_api_endpoint base URL, no trailing slash
# CLAUDE_PLUGIN_OPTION_api_token fmcp_ API key (sensitive)
# SCRIBE_URL / SCRIBE_TOKEN override for the settings.json dogfooding path.
set -uo pipefail
command -v jq >/dev/null 2>&1 || exit 0
command -v curl >/dev/null 2>&1 || exit 0
# PreToolUse delivers { session_id, cwd, tool_name, tool_input: {...}, ... }
event=$(cat 2>/dev/null || true)
file_path=$(printf '%s' "$event" | jq -r '.tool_input.file_path // empty' 2>/dev/null) || exit 0
session_id=$(printf '%s' "$event" | jq -r '.session_id // empty' 2>/dev/null) || session_id=""
event_cwd=$(printf '%s' "$event" | jq -r '.cwd // empty' 2>/dev/null) || event_cwd=""
[ -n "$file_path" ] || exit 0
# The code about to be written. Write and Edit name this field differently, and
# the names have changed across Claude Code versions — take whichever is present
# rather than betting on one shape.
code=$(printf '%s' "$event" | jq -r '
.tool_input.content // .tool_input.file_content //
.tool_input.new_string // .tool_input.new_str // empty' 2>/dev/null) || code=""
# Skip formats that hold prose or data rather than reusable code. Purely to
# avoid a pointless round-trip — the server would return nothing for these
# anyway. Config formats are NOT skipped: a CI workflow or a compose file is
# often exactly the thing worth reusing.
case "$file_path" in
*.md|*.mdx|*.txt|*.rst|*.json|*.lock|*.log|*.csv|*.tsv|*.svg|*.png|*.jpg|*.jpeg|*.gif|*.ico|*.pdf)
exit 0 ;;
esac
url=${SCRIBE_URL:-${CLAUDE_PLUGIN_OPTION_api_endpoint:-}}
token=${SCRIBE_TOKEN:-${CLAUDE_PLUGIN_OPTION_api_token:-}}
# Guard against an unexpanded ${...} placeholder arriving as a literal.
case "$url" in *'${'*) url="" ;; esac
case "$token" in *'${'*) token="" ;; esac
# Unconfigured install → silent. Prior-art recall is pure enrichment.
[ -n "$url" ] && [ -n "$token" ] || exit 0
# Snippet locations are recorded repo-relative, so send a repo-relative path —
# an absolute one would simply match nothing.
lookup_dir=$(dirname -- "$file_path" 2>/dev/null || true)
[ -d "$lookup_dir" ] || lookup_dir=${event_cwd:-${CLAUDE_PROJECT_DIR:-$PWD}}
repo_root=$(git -C "$lookup_dir" rev-parse --show-toplevel 2>/dev/null || true)
rel_path="$file_path"
if [ -n "$repo_root" ]; then
case "$file_path" in
"$repo_root"/*) rel_path="${file_path#"$repo_root"/}" ;;
esac
fi
# Cap the code sent as the semantic query. The embedder truncates at its own
# token limit well before this, so a bigger slice buys no extra signal — and the
# payload has to stay a GET (a read-scoped API key cannot POST, and every other
# plugin hook works with a read key).
q=$(printf '%s' "$code" | cut -c1-1200)
path_enc=$(printf '%s' "$rel_path" | jq -rR '@uri' 2>/dev/null) || exit 0
code_enc=$(printf '%s' "$q" | jq -rR '@uri' 2>/dev/null) || code_enc=""
# Resolve the working repo's remote so the server can scope to the bound project.
repo=$(git -C "$lookup_dir" remote get-url origin 2>/dev/null || true)
repo_q=""
if [ -n "$repo" ]; then
enc=$(printf '%s' "$repo" | jq -rR '@uri' 2>/dev/null) || enc=""
[ -n "$enc" ] && repo_q="&repo=${enc}"
fi
# Per-session dedup, in its own file rather than sharing auto-inject's. Each
# surface shows a given snippet at most once per session, but they don't silence
# each other: a title that flew past in a prompt menu twenty turns ago is
# exactly what should reappear at the moment the duplicate is being written.
state_dir="${TMPDIR:-/tmp}/scribe-priorart"
mkdir -p "$state_dir" 2>/dev/null || true
idfile=""
exclude_q=""
if [ -n "$session_id" ]; then
safe_sid=$(printf '%s' "$session_id" | tr -c 'A-Za-z0-9._-' '_')
idfile="$state_dir/${safe_sid}.ids"
if [ -f "$idfile" ]; then
seen=$(tr '\n' ',' < "$idfile" 2>/dev/null | sed 's/,$//')
[ -n "$seen" ] && exclude_q="&exclude_ids=${seen}"
fi
fi
body=$(curl -fsS --max-time 5 \
-H "Authorization: Bearer ${token}" \
"${url%/}/api/plugin/prior-art?path=${path_enc}&code=${code_enc}${repo_q}${exclude_q}" 2>/dev/null) || exit 0
[ -n "$body" ] || exit 0
context=$(printf '%s' "$body" | jq -r '.context // empty' 2>/dev/null) || exit 0
[ -n "$context" ] || exit 0
# Remember what was surfaced so it isn't shown again this session.
if [ -n "$idfile" ]; then
printf '%s' "$body" | jq -r '.note_ids[]? // empty' 2>/dev/null >> "$idfile" || true
fi
# No permissionDecision: this is a nudge, not a gate. The write goes ahead.
jq -n --arg c "$context" \
'{hookSpecificOutput: {hookEventName: "PreToolUse", additionalContext: $c}}'
exit 0
+5
View File
@@ -32,6 +32,11 @@ through recall/auto-inject; this skill is the active reflex around that.
`location` points at the reference implementation. Adapt, don't re-derive.
- If auto-inject already surfaced a snippet title that looks relevant, that's
your cue to `get_snippet` it rather than start from scratch.
- **Prior art offered beside a write is not noise — read it.** When Scribe notes
that a snippet is already recorded for the file you just wrote or edited, open
it before you go any further. Either it's the helper you were about to
duplicate — reuse it and drop yours — or it isn't, and the record needs the new
location adding. Both are cheaper now than after the duplicate settles in.
## The moment you build something reusable — record it
+47
View File
@@ -99,6 +99,53 @@ async def autoinject_retrieve():
return jsonify(result)
@plugin_bp.get("/prior-art")
@login_required
async def write_path_prior_art():
"""Prior-art hint for the plugin's PreToolUse hook on Write/Edit.
Answers "what is already recorded for the file about to be written?" — by
place (a snippet at this path or in its directory) and by meaning (snippets
resembling the code about to be written). Titles only, gated, and empty most
of the time. See services.plugin_context.build_write_path_hint.
Query:
path (str) — the target file, REPO-RELATIVE, matching the
convention snippet locations are recorded in. An
absolute path simply won't match anything.
code (optional) — the code about to be written; the semantic query.
Omit it and only the location arm runs.
repo (optional) — working repo remote; resolved to the bound project
to scope the search, exactly as /retrieve does. It
is NOT used as the location `repo` filter — see the
service docstring for why those two differ.
project_id (opt) — explicit project scope override (ad-hoc/testing).
exclude_ids (opt) — comma-separated ids already surfaced this session.
"""
path = (request.args.get("path") or "").strip()
code = request.args.get("code") or ""
try:
project_id = int(request.args.get("project_id", 0) or 0)
except (TypeError, ValueError):
project_id = 0
repo = (request.args.get("repo") or "").strip()
if repo and not project_id:
resolved = await repo_bindings_svc.resolve_project(g.user.id, repo)
if resolved:
project_id = resolved
exclude_ids = [
int(p) for p in (request.args.get("exclude_ids") or "").split(",")
if p.strip().isdigit()
]
result = await plugin_ctx_svc.build_write_path_hint(
g.user.id, path, code=code, project_id=project_id, exclude_ids=exclude_ids
)
return jsonify(result)
@plugin_bp.get("/processes")
@login_required
async def process_manifest():
+9
View File
@@ -114,6 +114,7 @@ async def semantic_search_notes(
threshold: float = _SIMILARITY_THRESHOLD,
project_id: int | None = None,
is_task: bool | None = None,
note_type: str | None = None,
orphan_only: bool = False,
scope: str = "own",
) -> list[tuple[float, Note]]:
@@ -122,6 +123,9 @@ async def semantic_search_notes(
Scores are cosine similarities in [-1, 1]; only notes at or above
*threshold* are returned, sorted highest-first.
`note_type` narrows to a single record kind (e.g. "snippet"), for callers
that want prior art rather than everything embedded.
`scope` ("own" | "browse" | "read", see access.notes_visibility_clause)
decides how far this may see. It exists because this one function serves
three different kinds of act: an explicit search, which should reach
@@ -175,6 +179,11 @@ async def semantic_search_notes(
stmt = stmt.where(Note.status.isnot(None))
elif is_task is False:
stmt = stmt.where(Note.status.is_(None))
# Narrow to one kind of record. Composes with is_task rather than
# replacing it — 'snippet' is a non-task note_type, so a caller asking
# for prior art gets snippets and not the dev-log that mentions them.
if note_type:
stmt = stmt.where(Note.note_type == note_type)
if exclude_ids:
stmt = stmt.where(NoteEmbedding.note_id.notin_(exclude_ids))
stmt = stmt.where(distance <= max_distance).order_by(distance.asc()).limit(limit)
+183
View File
@@ -14,6 +14,7 @@ index alone already steers behavior.
"""
from __future__ import annotations
import logging
import re
import time
@@ -25,11 +26,14 @@ from scribe.services import knowledge as knowledge_svc
from scribe.services import notes as notes_svc
from scribe.services import projects as projects_svc
from scribe.services import rulebooks as rulebooks_svc
from scribe.services import snippets as snippets_svc
from scribe.services.access import label_shared_items, owner_names_for
from scribe.services.embeddings import semantic_search_notes
from scribe.services.retrieval_telemetry import record_retrieval
from scribe.services.settings import get_setting
logger = logging.getLogger(__name__)
# Defensive cap below Claude Code's 10k additionalContext limit.
_MAX_CHARS = 9000
@@ -51,6 +55,17 @@ AUTOINJECT_DEFAULT_ENABLED = True
AUTOINJECT_DEFAULT_THRESHOLD = 0.55
AUTOINJECT_DEFAULT_TOP_K = 3
# The write-path trigger (#2082) gets its own on/off switch but SHARES the
# threshold and top-k above. One knob for "how loud may Scribe be" is easier to
# reason about than two that drift; the switch is separate because wanting prior
# art on writes while keeping prompts quiet (or the reverse) is a real
# preference, and it's the only part of the gate an operator can judge without
# data. The two surfaces log under different `source` values, so if the
# telemetry ever shows they want different thresholds, splitting them is a
# data-backed change rather than a guess made up front.
WRITEPATH_ENABLED_KEY = "kb_writepath_enabled"
WRITEPATH_DEFAULT_ENABLED = True
# Margin gate: drop any hit more than this far below the top hit's score, so a
# single strong match doesn't drag in a wall of barely-passing neighbours.
_AUTOINJECT_BAND = 0.10
@@ -259,6 +274,174 @@ async def build_autoinject_hint(
return {"context": "\n".join(lines), "note_ids": note_ids, "config": cfg}
# --- Write-path trigger (#2082): prior art at the moment code is written ------
# Auto-inject above fires on the operator's prompt. The moment reuse is actually
# lost is later — when the AGENT decides mid-task to write a helper — and nothing
# fired there. This is that trigger: the plugin's PreToolUse hook on Write/Edit
# asks what prior art is already recorded for the file being written.
#
# Two arms, deliberately different in kind:
# - BY PLACE — a snippet recorded at this path (or in its directory) is prior
# art by definition, not by resemblance, so it isn't scored or thresholded.
# This is what the reverse lookup (#2083) was built to answer.
# - BY MEANING — semantic search over snippets only, using the code about to be
# written, under the same gates as auto-inject.
# Place beats meaning in the menu because "there is already a canonical helper in
# this exact file" is a stronger claim than "this resembles something".
def _prior_art_line(item: dict, marker: str, owner: str | None) -> str:
"""One menu line: `- #12 [here] "title"`, attributed when it isn't yours."""
title = (item.get("title") or "(untitled)").replace("\n", " ").strip()
line = f"> - #{item['id']} [{marker}] \"{title}\""
if owner:
line += f" — shared by {owner}, treat as a suggestion"
return line
async def get_writepath_config(user_id: int) -> dict:
"""Write-path trigger settings: its own `enabled`, auto-inject's gates."""
cfg = await get_autoinject_config(user_id)
enabled_raw = await get_setting(
user_id, WRITEPATH_ENABLED_KEY,
"true" if WRITEPATH_DEFAULT_ENABLED else "false",
)
return {
**cfg,
"enabled": enabled_raw.strip().lower() in ("true", "1", "yes", "on"),
}
async def build_write_path_hint(
user_id: int,
path: str,
code: str = "",
project_id: int = 0,
exclude_ids: list[int] | None = None,
) -> dict:
"""Prior-art hint for the plugin's PreToolUse hook on Write/Edit.
`path` is the file about to be written, REPO-RELATIVE — matching the
convention snippet locations are recorded in. `code` is what's about to be
written, used only as the semantic query.
Same four anti-bloat gates as auto-inject (threshold, margin, session dedup
via `exclude_ids`, titles-never-bodies) plus the shared top-k cap across BOTH
arms — so a file with a lot of recorded history can't turn one edit into a
wall of text. Returns empty context when disabled, when there's no path, or
when nothing is recorded — which is the common case, and the point.
Note the repo↔project mapping is deliberately one-way: the hook sends a git
remote, which the ROUTE resolves to `project_id` through the repo bindings.
It is never used as the location `repo` filter — a snippet's `repo` is a
free-text label the operator typed ("Scribe"), not a remote URL, and matching
one against the other would silently return nothing.
Returns {"context": str, "note_ids": list[int], "config": dict}. The semantic
arm is logged to retrieval_logs as source='write_path' — its own source, so
its precision is tunable separately from auto-inject's.
KNOWN GAP: only the semantic arm is logged, matching auto-inject's convention
of recording the candidate set for threshold tuning. Location hits carry no
score, so folding them in would corrupt the score distribution the log exists
to capture — but it does mean a snippet surfaced BY PLACE leaves no trace. The
usage signal (#2085) correlates retrieval_logs against later pulls, so it will
need a home for un-scored surfacing before it can measure this arm.
"""
cfg = await get_writepath_config(user_id)
empty = {"context": "", "note_ids": [], "config": cfg}
path = (path or "").strip()
if not cfg["enabled"] or not path:
return empty
top_k = cfg["top_k"]
excluded = set(exclude_ids or [])
scope_project = project_id or None
# --- arm 1: by place ---
here: list[dict] = []
nearby: list[dict] = []
try:
here, _ = await snippets_svc.list_snippets(
user_id, path=path, limit=top_k, project_id=scope_project,
)
directory = path.rsplit("/", 1)[0] if "/" in path else ""
if directory and len(here) < top_k:
nearby, _ = await snippets_svc.list_snippets(
user_id, path=directory, limit=top_k, project_id=scope_project,
)
except Exception:
logger.warning("Write-path location lookup failed", exc_info=True)
seen: set[int] = set(excluded)
placed: list[tuple[str, dict]] = []
for marker, items in (("here", here), ("nearby", nearby)):
for item in items:
nid = int(item["id"])
if nid in seen:
continue
seen.add(nid)
placed.append((marker, item))
# --- arm 2: by meaning ---
scored: list[tuple[str, dict]] = []
remaining = top_k - len(placed)
query = (code or "").strip()
if remaining > 0 and query:
t0 = time.perf_counter()
hits = await semantic_search_notes(
user_id, query,
limit=remaining,
threshold=cfg["threshold"],
project_id=scope_project,
exclude_ids=seen,
note_type="snippet",
# Same reasoning as auto-inject: nobody asked for this, so it takes
# the browse scope and never surfaces a one-to-one direct share.
scope="browse",
)
record_retrieval(
user_id=user_id, source="write_path", query=query,
threshold=cfg["threshold"], limit=remaining,
project_id=scope_project, is_task=False, results=hits,
duration_ms=(time.perf_counter() - t0) * 1000.0,
)
if hits:
top_score = hits[0][0]
for score, note in hits:
if score < top_score - _AUTOINJECT_BAND:
continue
scored.append((
f"similar {score:.2f}",
{"id": int(note.id), "title": note.title, "user_id": note.user_id},
))
menu = (placed + scored)[:top_k]
if not menu:
return empty
owners = await owner_names_for({
int(it["user_id"]) for _m, it in menu
if it.get("user_id") is not None and int(it["user_id"]) != user_id
})
lines = [
f"> Prior art already recorded in Scribe for `{path}` — open one with "
"`get_snippet(id)` and reuse it rather than writing a fresh one-off "
"(titles only; shown once per session):",
]
note_ids: list[int] = []
for marker, item in menu:
note_ids.append(int(item["id"]))
owner_id = item.get("user_id")
owner = None
if owner_id is not None and int(owner_id) != user_id:
owner = owners.get(int(owner_id)) or "another user"
lines.append(_prior_art_line(item, marker, owner))
return {"context": "\n".join(lines), "note_ids": note_ids, "config": cfg}
async def _topic_titles(topic_ids: set[int]) -> dict[int, str]:
"""Map topic_id -> title for the given ids (live topics only)."""
if not topic_ids:
+405
View File
@@ -0,0 +1,405 @@
"""Tests for the write-path trigger (#2082) — prior art at the moment code is written.
Covers the service (`build_write_path_hint`): the two arms and their precedence,
the anti-bloat gates carried over from milestone 93, telemetry under its own
source, and the two ways this must stay silent. Plus the plugin hook contract —
a PreToolUse hook that returns a permission decision would be able to block the
operator's edit, which this feature must never do.
"""
import json
import re
import subprocess
from pathlib import Path
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
PLUGIN = Path(__file__).resolve().parents[1] / "plugin"
HOOK = PLUGIN / "hooks" / "scribe_prior_art.sh"
def _snippet_item(nid, title, user_id=1):
"""A row as services.snippets.list_snippets returns it (a dict, not a Note)."""
return {"id": nid, "title": title, "user_id": user_id, "note_type": "snippet"}
def _note(nid, title, user_id=1):
n = MagicMock()
n.id, n.title, n.user_id = nid, title, user_id
return n
def _cfg(**over):
base = {"enabled": True, "threshold": 0.55, "top_k": 3}
base.update(over)
return base
# --- silence: the common case ------------------------------------------------
@pytest.mark.asyncio
async def test_disabled_returns_empty_and_touches_nothing():
from scribe.services import plugin_context as pc
search, listing = AsyncMock(), AsyncMock()
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg(enabled=False))), \
patch.object(pc.snippets_svc, "list_snippets", listing), \
patch.object(pc, "semantic_search_notes", search):
out = await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
assert out["context"] == "" and out["note_ids"] == []
listing.assert_not_called()
search.assert_not_called()
@pytest.mark.asyncio
async def test_no_path_returns_empty():
"""Nothing to look up by place, and the semantic arm alone isn't this feature."""
from scribe.services import plugin_context as pc
search = AsyncMock()
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc, "semantic_search_notes", search):
out = await pc.build_write_path_hint(1, " ", code="def f(): ...")
assert out["context"] == ""
search.assert_not_called()
@pytest.mark.asyncio
async def test_nothing_recorded_is_silent():
from scribe.services import plugin_context as pc
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets", AsyncMock(return_value=([], 0))), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=[])), \
patch.object(pc, "record_retrieval", MagicMock()):
out = await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
assert out["context"] == "" and out["note_ids"] == []
# --- arm 1: by place ---------------------------------------------------------
@pytest.mark.asyncio
async def test_snippet_at_this_path_is_surfaced_without_a_score():
"""A snippet recorded HERE is prior art by definition, not by resemblance —
it must not be subject to the similarity threshold."""
from scribe.services import plugin_context as pc
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets",
AsyncMock(return_value=([_snippet_item(12, "debounce — rate-limit a callback")], 1))), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=[])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
assert out["note_ids"] == [12]
assert '#12 [here] "debounce — rate-limit a callback"' in out["context"]
assert "get_snippet(id)" in out["context"]
assert "src/x.py" in out["context"]
@pytest.mark.asyncio
async def test_directory_is_only_consulted_when_the_file_leaves_room():
from scribe.services import plugin_context as pc
calls = []
async def _listing(uid, **kw):
calls.append(kw["path"])
if kw["path"] == "src/lib/x.py":
return ([_snippet_item(i, f"s{i}") for i in (1, 2, 3)], 3)
return ([_snippet_item(9, "nearby")], 1)
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg(top_k=3))), \
patch.object(pc.snippets_svc, "list_snippets", _listing), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=[])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/lib/x.py", code="x")
# The file alone filled the budget, so the directory was never queried.
assert calls == ["src/lib/x.py"]
assert out["note_ids"] == [1, 2, 3]
@pytest.mark.asyncio
async def test_directory_hits_are_marked_nearby_not_here():
from scribe.services import plugin_context as pc
async def _listing(uid, **kw):
if kw["path"] == "src/lib":
return ([_snippet_item(9, "sibling helper")], 1)
return ([], 0)
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets", _listing), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=[])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/lib/x.py", code="x")
assert '#9 [nearby] "sibling helper"' in out["context"]
@pytest.mark.asyncio
async def test_a_failing_location_lookup_does_not_sink_the_hint():
"""The place arm is best-effort — a broken query must not cost the semantic one."""
from scribe.services import plugin_context as pc
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets", AsyncMock(side_effect=RuntimeError("boom"))), \
patch.object(pc, "semantic_search_notes",
AsyncMock(return_value=[(0.80, _note(5, "throttle — …"))])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
assert out["note_ids"] == [5]
# --- arm 2: by meaning, under the milestone-93 gates -------------------------
@pytest.mark.asyncio
async def test_semantic_arm_is_snippet_only_and_browse_scoped():
from scribe.services import plugin_context as pc
search = AsyncMock(return_value=[])
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets", AsyncMock(return_value=([], 0))), \
patch.object(pc, "semantic_search_notes", search), \
patch.object(pc, "record_retrieval", MagicMock()):
await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
kwargs = search.await_args.kwargs
# Prior art means snippets — not the dev-log that happens to mention one.
assert kwargs["note_type"] == "snippet"
# Nobody asked for this, so it must not reach a one-to-one direct share.
assert kwargs["scope"] == "browse"
@pytest.mark.asyncio
async def test_margin_gate_applies_to_the_semantic_arm():
from scribe.services import plugin_context as pc
hits = [(0.80, _note(11, "near")), (0.74, _note(22, "alsoNear")), (0.61, _note(33, "far"))]
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg(top_k=5))), \
patch.object(pc.snippets_svc, "list_snippets", AsyncMock(return_value=([], 0))), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=hits)), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/x.py", code="def f(): ...")
assert out["note_ids"] == [11, 22]
assert "#33" not in out["context"]
@pytest.mark.asyncio
async def test_no_code_means_no_semantic_arm():
"""An Edit whose payload we couldn't read still gets the location answer."""
from scribe.services import plugin_context as pc
search = AsyncMock()
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets",
AsyncMock(return_value=([_snippet_item(12, "here helper")], 1))), \
patch.object(pc, "semantic_search_notes", search), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/x.py", code="")
assert out["note_ids"] == [12]
search.assert_not_called()
@pytest.mark.asyncio
async def test_place_beats_meaning_and_the_cap_covers_both_arms():
from scribe.services import plugin_context as pc
async def _listing(uid, **kw):
if kw["path"] == "src/x.py":
return ([_snippet_item(1, "placed")], 1)
return ([], 0)
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg(top_k=2))), \
patch.object(pc.snippets_svc, "list_snippets", _listing), \
patch.object(pc, "semantic_search_notes",
AsyncMock(return_value=[(0.9, _note(7, "scored"))])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
out = await pc.build_write_path_hint(1, "src/x.py", code="x")
# Location hit first; the whole menu stays within top_k.
assert out["note_ids"] == [1, 7]
assert len(out["note_ids"]) <= 2
assert out["context"].index("#1") < out["context"].index("#7")
@pytest.mark.asyncio
async def test_semantic_arm_only_asks_for_the_budget_the_place_arm_left():
from scribe.services import plugin_context as pc
search = AsyncMock(return_value=[])
async def _listing(uid, **kw):
if kw["path"] == "src/x.py":
return ([_snippet_item(1, "placed")], 1)
return ([], 0)
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg(top_k=3))), \
patch.object(pc.snippets_svc, "list_snippets", _listing), \
patch.object(pc, "semantic_search_notes", search), \
patch.object(pc, "record_retrieval", MagicMock()):
await pc.build_write_path_hint(1, "src/x.py", code="x")
assert search.await_args.kwargs["limit"] == 2
@pytest.mark.asyncio
async def test_session_dedup_excludes_ids_from_both_arms():
from scribe.services import plugin_context as pc
search = AsyncMock(return_value=[])
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets",
AsyncMock(return_value=([_snippet_item(12, "already shown")], 1))), \
patch.object(pc, "semantic_search_notes", search), \
patch.object(pc, "record_retrieval", MagicMock()):
out = await pc.build_write_path_hint(1, "src/x.py", code="x", exclude_ids=[12])
# The location hit was already surfaced this session → dropped, not repeated.
assert out["note_ids"] == []
assert 12 in search.await_args.kwargs["exclude_ids"]
# --- attribution + telemetry -------------------------------------------------
@pytest.mark.asyncio
async def test_another_users_snippet_is_attributed():
"""Rule #47 / the #2084 lesson: an unattributed line reads as your own record."""
from scribe.services import plugin_context as pc
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets",
AsyncMock(return_value=([_snippet_item(12, "theirs", user_id=42)], 1))), \
patch.object(pc, "semantic_search_notes", AsyncMock(return_value=[])), \
patch.object(pc, "record_retrieval", MagicMock()), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={42: "alex"})):
out = await pc.build_write_path_hint(1, "src/x.py", code="x")
assert "shared by alex, treat as a suggestion" in out["context"]
@pytest.mark.asyncio
async def test_telemetry_uses_its_own_source():
"""Separate source is what lets this surface be tuned apart from auto_inject."""
from scribe.services import plugin_context as pc
rec = MagicMock()
with patch.object(pc, "get_writepath_config", AsyncMock(return_value=_cfg())), \
patch.object(pc.snippets_svc, "list_snippets", AsyncMock(return_value=([], 0))), \
patch.object(pc, "semantic_search_notes",
AsyncMock(return_value=[(0.9, _note(7, "scored"))])), \
patch.object(pc, "record_retrieval", rec), \
patch.object(pc, "owner_names_for", AsyncMock(return_value={})):
await pc.build_write_path_hint(1, "src/x.py", code="x", project_id=4)
rec.assert_called_once()
assert rec.call_args.kwargs["source"] == "write_path"
assert rec.call_args.kwargs["source"] != "auto_inject"
assert rec.call_args.kwargs["project_id"] == 4
@pytest.mark.asyncio
async def test_config_has_its_own_switch_but_shares_the_gates():
from scribe.services import plugin_context as pc
stored = {pc.WRITEPATH_ENABLED_KEY: "false"}
with patch.object(pc, "get_setting",
AsyncMock(side_effect=lambda uid, k, d: stored.get(k, d))):
cfg = await pc.get_writepath_config(1)
# Its own switch is off while auto-inject stays on...
assert cfg["enabled"] is False
# ...and the gates are auto-inject's, not a second set that could drift.
assert cfg["threshold"] == pc.AUTOINJECT_DEFAULT_THRESHOLD
assert cfg["top_k"] == pc.AUTOINJECT_DEFAULT_TOP_K
# --- the route between them --------------------------------------------------
def test_route_reads_every_arg_the_hook_sends():
"""Rule #33: the hook's query string and the route's reader are one contract,
and the hook is the only caller — a rename on either side is silent."""
import inspect
from scribe.routes import plugin as routes
src = inspect.getsource(routes.write_path_prior_art)
for arg in ("path", "code", "repo", "project_id", "exclude_ids"):
assert f'request.args.get("{arg}"' in src, f"route ignores {arg}"
hook = HOOK.read_text()
for arg in ("path=", "code=", "repo=", "exclude_ids="):
assert arg in hook, f"hook never sends {arg}"
def test_route_resolves_repo_to_a_project_not_to_a_location_filter():
"""A snippet's `repo` is a label the operator typed; the hook sends a git
remote. Matching one against the other would silently return nothing, so the
remote may only reach resolve_project."""
import inspect
from scribe.routes import plugin as routes
src = inspect.getsource(routes.write_path_prior_art)
assert "resolve_project" in src
assert "repo=repo" not in src
# --- the plugin surface ------------------------------------------------------
def test_hook_is_registered_for_write_and_edit():
cfg = json.loads((PLUGIN / "hooks" / "hooks.json").read_text())
entries = cfg["hooks"]["PreToolUse"]
assert any(
"Write" in e.get("matcher", "") and "Edit" in e.get("matcher", "")
and any("scribe_prior_art.sh" in h["command"] for h in e["hooks"])
for e in entries
)
def test_hook_never_returns_a_permission_decision():
"""The load-bearing property: this hook informs a write, it cannot stop one.
A permissionDecision of deny/ask would put a recall aid in the way of the
operator's work."""
src = HOOK.read_text()
assert "additionalContext" in src
code_lines = [ln for ln in src.splitlines() if not ln.lstrip().startswith("#")]
assert not any("permissionDecision" in ln for ln in code_lines)
def test_hook_is_executable_and_shell_valid():
assert HOOK.stat().st_mode & 0o111, "hook must be executable"
subprocess.run(["bash", "-n", str(HOOK)], check=True)
def test_hook_reads_both_write_and_edit_payload_shapes():
"""Write and Edit name the payload differently, and the names have changed
across Claude Code versions — read whichever is present."""
src = HOOK.read_text()
for field in ("content", "file_content", "new_string", "new_str"):
assert f".tool_input.{field}" in src
assert ".tool_input.file_path" in src
def test_hook_stays_a_get_so_a_read_scoped_key_works():
"""Every other plugin hook works with a read-scoped key; a POST would demand
write scope (see auth.py) and silently break those installs."""
src = HOOK.read_text()
assert "-X POST" not in src and "--data" not in src
assert "/api/plugin/prior-art?" in src
def test_hook_exits_silently_when_unconfigured():
"""An install with no Scribe URL/token must produce no output at all."""
out = subprocess.run(
["bash", str(HOOK)],
input=json.dumps({
"session_id": "s1", "cwd": "/tmp", "tool_name": "Write",
"tool_input": {"file_path": "/tmp/x.py", "content": "def f(): ..."},
}),
capture_output=True, text=True,
env={"PATH": "/usr/bin:/bin", "SCRIBE_URL": "", "SCRIBE_TOKEN": ""},
)
assert out.returncode == 0
assert out.stdout.strip() == ""
def test_hook_skips_prose_and_data_files():
"""No round-trip for a markdown edit — the server would return nothing anyway."""
src = HOOK.read_text()
skip = re.search(r"case \"\$file_path\" in\n(.*?)esac", src, re.S)
assert skip, "expected an extension skip list"
for ext in ("*.md", "*.json", "*.lock", "*.png"):
assert ext in skip.group(1)
# Config formats are deliberately NOT skipped — a workflow file is reusable.
assert "*.yml" not in skip.group(1)
def test_plugin_version_bumped_with_the_hook():
"""The #1040 lesson: a plugin change clients can't see is a change that didn't
ship."""
manifest = json.loads((PLUGIN / ".claude-plugin" / "plugin.json").read_text())
version = tuple(int(p) for p in manifest["version"].split("."))
assert version >= (0, 1, 18)