@onlooker-community/ecosystem 0.34.0 → 0.34.1

package/.claude/skills/writing-tests/SKILL.md

@@ -160,6 +160,33 @@ _seed_artifact() {
 }
 ```
 
+## Gate every assertion: non-final `[[ ]]` needs `|| return 1`
+
+`bats` resolves `#!/usr/bin/env bash`, which on macOS is the system bash 3.2.
+Under 3.2 a failing **non-final** `[[ ]]` does **not** fail the test body —
+execution continues and the test reports `ok`. Single-bracket `[ ]`, failing
+commands, and failing command substitutions all gate correctly even there; the
+defect is specific to `[[ ]]`. Under bash 5.x (what CI runs) it gates properly,
+so a broken assertion passes locally and only CI would catch it.
+
+Only the **last** assertion in a test body is safe, because a body's exit
+status is its last command. Any `[[ ]]` before that needs explicit gating:
+
+```bash
+@test "example" {
+  [[ "$output" == *"expected"* ]] || return 1   # non-final — must gate
+  [ "$status" -eq 0 ]                            # final — gates on its own
+}
+```
+
+Do **not** "fix" this by swapping to `[ ]`. Most of these assertions rely on
+`[[ ]]`-only behavior — `==` glob patterns, `=~` regex, internal `&&`/`||` —
+and a blanket swap silently changes what they test.
+
+When you add an assertion, break it on purpose once and confirm the test fails.
+A test that passes whether or not the code is correct is worse than no test,
+because it reports coverage that does not exist.
+
 ## Anti-patterns
 
 Don't hand-roll these — each has bitten the suite before:

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "ecosystem",
-  "version": "0.34.0",
+  "version": "0.34.1",
   "description": "Observability substrate for Claude Code. Provides the shared $ONLOOKER_DIR storage root (default $HOME/.onlooker), canonical schema-validated event emission, session and tool tracking hooks, and prompt rules. Required by all other Onlooker plugins.",
   "author": {
     "name": "Onlooker Community",

package/.release-please-manifest.json

@@ -1,7 +1,7 @@
 {
-  ".": "0.34.0",
+  ".": "0.34.1",
   "plugins/archivist": "0.3.1",
-  "plugins/tribunal": "1.1.2",
+  "plugins/tribunal": "1.1.3",
   "plugins/echo": "0.3.1",
   "plugins/cartographer": "0.3.1",
   "plugins/governor": "0.3.2",
@@ -9,10 +9,10 @@
   "plugins/scribe": "0.4.2",
   "plugins/counsel": "0.5.1",
   "plugins/warden": "0.3.1",
-  "plugins/librarian": "0.7.0",
-  "plugins/curator": "0.2.2",
-  "plugins/historian": "0.3.2",
-  "plugins/assayer": "1.1.2",
+  "plugins/librarian": "0.7.1",
+  "plugins/curator": "0.2.3",
+  "plugins/historian": "0.3.3",
+  "plugins/assayer": "1.1.3",
   "plugins/bursar": "0.3.0",
   "plugins/lineage": "0.2.2",
   "plugins/inspector": "0.3.2"

package/AGENTS.md

@@ -47,6 +47,117 @@ cp -rf source dest          # NOT: cp -r source dest
 - `apt-get` - use `-y` flag
 - `brew` - use `HOMEBREW_NO_AUTO_UPDATE=1` env var
 
+<!-- The sections below are mirrored from CLAUDE.md. The two files are
+     independent — not symlinked, not sharing an inode — so an edit to the
+     repository's conventions must land in both or agents get different
+     instructions depending on which file their harness reads. Claude Code
+     reads CLAUDE.md; Codex and several other tools read this one. -->
+
+## Repository layout
+
+```
+ecosystem/                    ← substrate plugin (always-on observability)
+  hooks/                      ← session, tool, and prompt hooks
+  scripts/lib/                ← shared bash helpers and the canonical event emitter
+  skills/                     ← user-invocable slash commands
+  config.json                 ← ecosystem defaults
+
+plugins/
+  archivist/                  ← session memory across context truncation
+  bursar/                     ← multi-session, per-project budget rollup (governor's cross-session view)
+  cartographer/               ← instruction-file auditor (CLAUDE.md, AGENTS.md, rules/)
+  compass/                    ← pre-write alignment gate (design phase)
+  echo/                       ← prompt-change regression detection
+  governor/                   ← resource governance and budget enforcement
+  inspector/                  ← per-edit lint and typecheck gate
+  lineage/                    ← per-change provenance ("why does this line exist?")
+  tribunal/                   ← multi-agent quality gate (Actor → Jury → Meta-Judge → Gate)
+
+docs/
+  architecture.md             ← how plugins compose and share the event bus
+  adr/                        ← ecosystem-level architectural decisions
+
+scripts/lib/onlooker-event.mjs  ← canonical event builder; all plugins route through this
+~/.onlooker/                    ← shared runtime storage (logs, plugin artifacts)
+```
+
+## Plugin map
+
+| Plugin | Hook surface | When it fires |
+|--------|-------------|---------------|
+| ecosystem | SessionStart/End, PreToolUse, PostToolUse, PostToolUseFailure, UserPromptSubmit, UserPromptExpansion, PreCompact, PostCompact, TaskCreated, TaskCompleted, WorktreeCreate, WorktreeRemove | Always — substrate |
+| archivist | PreCompact, SessionStart | Extracts decisions/dead-ends on compaction; reinjects at next SessionStart |
+| cartographer | SessionStart, PostToolUse (Write, Edit, MultiEdit) | Audits instruction files on session start and after instruction-file writes |
+| compass | PreToolUse (Write, Edit, MultiEdit, Bash) | Before any write — alignment check |
+| echo | Stop | Regression-tests prompt changes after each agent stop |
+| governor | SessionStart, PreToolUse (Task), PostToolUse (Task), Stop | Budget gates on subagent spawns; tracks spend per session |
+| tribunal | Stop + skill invocation | Post-task quality gate; also invokable via `/tribunal` |
+| warden | PostToolUse (WebFetch, Read), PreToolUse (Write, Edit, MultiEdit, Bash), SessionStart + skill invocation | Scans ingested content for injection; closes a content gate that blocks write-class tools until cleared via `/warden` |
+| assayer | Stop | Verifies the agent's final-message claims against actual command results in the transcript; advisory |
+| bursar | SessionStart, SessionEnd | Rolls each session's spend into a per-project ledger on SessionEnd; surfaces "this project burned $X this week" at SessionStart. Governor is per-session; bursar is the cross-session rollup |
+| lineage | PostToolUse (Edit, Write, MultiEdit) + skill invocation | Records per-change provenance (session_id/turn + redacted, size-capped snippets) into a per-project ledger; `/lineage <file>:<line>` answers "why does this line exist?" by joining records to historian transcripts to recover prompt context |
+| inspector | PostToolUse (Write, Edit, MultiEdit) | Per-edit verification: runs the project's configured lint + typecheck commands on just the touched file and emits `inspector.check.*` / `inspector.run.completed`. Surfaces issues to the agent for the next turn. Cheaper than the planned proctor (which runs the full project verify at Stop); complements assayer (which catches claims the agent makes without running anything) |
+
+Plugins communicate by emitting events to the JSONL log — they do not call each other directly. All plugins depend on the ecosystem substrate; no plugin depends on another plugin directly.
+
+## Compass plugin (design phase)
+
+Compass is the pre-write alignment gate. It has no implementation yet. Design lives in `plugins/compass/docs/design.md`.
+
+**What it does:** Fires on `PreToolUse` for write-class tools. Samples N=5 parallel Haiku evaluators to score intent clarity. Blocks when `confidence < 0.65 OR stddev > 0.20` and surfaces a clarification prompt.
+
+**Critical architectural decision (ADR-001):** The evaluator must see the **prior assistant turn** alongside the current context — not the current context alone. Evaluating a reply in isolation produces a systematic false-positive class: a user answering an agent's enumerated question ("the internal one") looks ambiguous without the question that prompted it.
+
+The pipeline is:
+
+```
+Trigger Gate → Transcript Reader → Symbolic Skip Layer → Sanitizer → N=5 Evaluators → Gate
+```
+
+- **Transcript reader** resolves `prior_assistant_turn` from `transcript_path` in the hook JSON payload (same field tribunal-stop-gate.sh reads). Reads one turn back from that file (already committed before `PreToolUse` fires — no timing-skew risk). If `transcript_path` is absent or unreadable, proceeds with an empty prior turn.
+- **Symbolic skip layer** short-circuits to `confident` when the prior turn is an enumerated question and the current context is an option reference, without an LLM call. Controlled by `skip_patterns.reply_to_question.enabled` (default `true`).
+- **Evaluator prompt** uses a structured pair: `<prior_assistant_turn>` and `<context_excerpt>` as separate XML-delimited slots. The convergence question is: *"Given the prior assistant turn as context, would two independent readers converge on the same interpretation of this write?"*
+
+See `plugins/compass/docs/adr/001-evaluate-prompts-in-context.md` for the full decision record.
+
+## Adding a new plugin
+
+1. Create `plugins/<name>/` with `.claude-plugin/plugin.json`, `config.json`, `hooks/hooks.json`.
+2. Use `scripts/lib/onlooker-event.mjs` for all event emission — never write directly to the JSONL log.
+3. Store runtime artifacts under `${ONLOOKER_DIR:-$HOME/.onlooker}/<name>/<project-key>/`. Always use `$ONLOOKER_DIR` — never hardcode `~/.onlooker` — so the test suite's isolated temp home is respected.
+4. Derive the project key via `tribunal_project_key` (or equivalent) — first 12 hex chars of SHA256(`remote:<origin-url>`), falling back to SHA256(`root:<repo-root>`) for repos without a remote. See `plugins/tribunal/scripts/lib/tribunal-project-key.sh`.
+5. Register event types in `@onlooker-community/schema` before emitting them. The runtime emitter is dependency-free and **fails open**: it validates against the schema package only when that package is resolvable (dev, CI, tests) and emits unconditionally otherwise, because installed marketplace plugins ship no `node_modules`. Schema drift is caught in CI against the published schemas at `schema.onlooker.dev`. See [ADR-005](docs/adr/005-runtime-emitter-fails-open.md).
+6. Fail-soft when `~/.onlooker/` is absent — plugins must not block a session they were not invited to.
+
+## Development
+
+```bash
+mise install          # installs all tools declared in mise.toml
+npm ci
+npm test              # bats + schema validation
+npm run test:ci       # shellcheck + bats + schema + lint
+```
+
+Tests use an isolated temp home; nothing writes to your real `~/.onlooker/`.
+
+## Git workflow
+
+**Always open a PR — never push directly to `main`.** Even though bypass rights allow direct pushes, this repo uses release-please for automated changelogs and versioning, so every change must travel through a PR to be picked up correctly. CI also runs on PRs before merge, catching failures before they land.
+
+Workflow:
+1. Create a feature branch: `git switch -c <type>/<short-description>`
+2. Commit using `/commit`
+3. Push the branch and open a PR using `/git-workflow:pr`
+4. Wait for CI to pass before merging
+
+## Conventions
+
+- All hooks are bash scripts. No Python, no Node entry points in hook scripts (they may shell out to `node` for event emission or heavy lifting).
+- Hook scripts source shared helpers from `scripts/lib/` (or the plugin's own `scripts/lib/`).
+- Event types follow `<plugin>.<noun>.<verb>` — e.g. `compass.check.skipped`, `tribunal.gate.blocked`.
+- ULIDs everywhere for IDs (not UUIDs). Each plugin ships its own `*_ulid` helper (e.g. `archivist-ulid.sh`, `tribunal-ulid.sh`); there is no shared ecosystem helper. Copy `plugins/tribunal/scripts/lib/tribunal-ulid.sh` as a starting point and rename the function prefix.
+- Config defaults live in `config.json`. User overrides go in `~/.claude/settings.json` (global) or `.claude/settings.json` (per-project) under the plugin's namespace key (e.g. `"compass"`, `"tribunal"`). See ADR-004.
+
 <!-- BEGIN BEADS INTEGRATION v:1 profile:minimal hash:970c3bf2 -->
 ## Beads Issue Tracker
 

package/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [0.34.1](https://github.com/onlooker-community/ecosystem/compare/ecosystem-v0.34.0...ecosystem-v0.34.1) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [0.34.0](https://github.com/onlooker-community/ecosystem/compare/ecosystem-v0.33.1...ecosystem-v0.34.0) (2026-08-09)
 
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@onlooker-community/ecosystem",
-  "version": "0.34.0",
+  "version": "0.34.1",
   "description": "Agents, skills, hooks, commands, rules, and MCP configurations that power [Onlooker](https://onlooker.dev)",
   "author": {
     "name": "Onlooker Community",

package/plugins/assayer/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "assayer",
-  "version": "1.1.2",
+  "version": "1.1.3",
   "description": "Claim verification. At session end, parses the agent's final message for testable claims (\"I ran the tests, they pass\", \"the build is green\") and checks each against the actual command results in the session transcript, classifying it corroborated, contradicted, or unverifiable. Catches lying-without-malice. Advisory by default. Builds on the Onlooker ecosystem plugin.",
   "author": {
     "name": "Onlooker Community",

package/plugins/assayer/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [1.1.3](https://github.com/onlooker-community/ecosystem/compare/assayer-v1.1.2...assayer-v1.1.3) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [1.1.2](https://github.com/onlooker-community/ecosystem/compare/assayer-v1.1.1...assayer-v1.1.2) (2026-08-02)
 
 

package/plugins/assayer/scripts/lib/assayer-config.sh

@@ -57,3 +57,9 @@ assayer_config_timeout() {
 	v=$(assayer_config_get '.assayer.evaluation.timeout_seconds')
 	printf '%s' "${v:-60}"
 }
+
+assayer_config_final_message_chars() {
+	local v
+	v=$(assayer_config_get '.assayer.final_message_chars')
+	printf '%s' "${v:-6000}"
+}

package/plugins/curator/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "curator",
-  "version": "0.2.2",
+  "version": "0.2.3",
   "description": "Maintenance layer for the user's typed auto-memory store. At every SessionStart, runs four cheap heuristic checks (date_decayed, path_broken, broken_index, orphaned_memory) against the memories at ~/.claude/projects/<encoded>/memory/ inside a wall-clock budget. Surfaces findings as a one-line pointer to /curator review; never edits the memory store directly. Builds on the Onlooker ecosystem plugin.",
   "author": {
     "name": "Onlooker Community",

package/plugins/curator/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [0.2.3](https://github.com/onlooker-community/ecosystem/compare/curator-v0.2.2...curator-v0.2.3) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [0.2.2](https://github.com/onlooker-community/ecosystem/compare/curator-v0.2.1...curator-v0.2.2) (2026-08-02)
 
 

package/plugins/curator/scripts/lib/curator-emit.sh

@@ -39,7 +39,8 @@ _CURATOR_EVENT_JS="${_CURATOR_EVENT_JS:-$(_curator_resolve_event_js)}"
 curator_emit() {
 	local event_type="${1:-}"
 	local session_id="${2:-}"
-	local payload="${3:-{\}}"
+	local payload="${3:-}"
+	[ -z "$payload" ] && payload='{}'
 
 	[[ -z "$event_type" || -z "$session_id" ]] && return 0
 	[[ -z "$_CURATOR_EVENT_JS" || ! -f "$_CURATOR_EVENT_JS" ]] && return 0

package/plugins/historian/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "historian",
-  "version": "0.3.2",
+  "version": "0.3.3",
   "description": "Episodic memory layer. At SessionEnd, chunks and sanitizes the session transcript and stores chunks under $ONLOOKER_DIR/historian/<project-key>/sessions/ (default $HOME/.onlooker). On UserPromptSubmit, embeds the prompt and performs similarity retrieval over stored chunks to surface relevant past context. Builds on the Onlooker ecosystem plugin.",
   "author": {
     "name": "Onlooker Community",

package/plugins/historian/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [0.3.3](https://github.com/onlooker-community/ecosystem/compare/historian-v0.3.2...historian-v0.3.3) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [0.3.2](https://github.com/onlooker-community/ecosystem/compare/historian-v0.3.1...historian-v0.3.2) (2026-08-01)
 
 

package/plugins/historian/scripts/lib/historian-emit.sh

@@ -39,7 +39,8 @@ _HISTORIAN_EVENT_JS="${_HISTORIAN_EVENT_JS:-$(_historian_resolve_event_js)}"
 historian_emit() {
 	local event_type="${1:-}"
 	local session_id="${2:-}"
-	local payload="${3:-{\}}"
+	local payload="${3:-}"
+	[ -z "$payload" ] && payload='{}'
 
 	[[ -z "$event_type" || -z "$session_id" ]] && return 0
 	[[ -z "$_HISTORIAN_EVENT_JS" || ! -f "$_HISTORIAN_EVENT_JS" ]] && return 0

package/plugins/librarian/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "librarian",
-  "version": "0.7.0",
+  "version": "0.7.1",
   "description": "Consolidation layer between archivist's per-session artifacts and the user's durable typed memory store. Detects which session decisions, dead-ends, and open questions deserve to live across sessions, classifies them into the user/feedback/project/reference types, and queues them as proposals for explicit confirmation. Auto-promotion is opt-in. Builds on the Onlooker ecosystem plugin.",
   "author": {
     "name": "Onlooker Community",

package/plugins/librarian/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [0.7.1](https://github.com/onlooker-community/ecosystem/compare/librarian-v0.7.0...librarian-v0.7.1) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [0.7.0](https://github.com/onlooker-community/ecosystem/compare/librarian-v0.6.1...librarian-v0.7.0) (2026-08-09)
 
 

package/plugins/librarian/scripts/lib/librarian-cli.sh

@@ -117,7 +117,8 @@ _librarian_cli_set_proposal_status() {
 	local key="$1"
 	local proposal_id="$2"
 	local new_status="$3"
-	local extra_json="${4:-{\}}"
+	local extra_json="${4:-}"
+	[ -z "$extra_json" ] && extra_json='{}'
 	[[ -z "$key" || -z "$proposal_id" || -z "$new_status" ]] && return 1
 
 	local path

package/plugins/librarian/scripts/lib/librarian-emit.sh

@@ -50,7 +50,8 @@ _LIBRARIAN_EVENT_JS="${_LIBRARIAN_EVENT_JS:-$(_librarian_resolve_event_js)}"
 librarian_emit() {
 	local event_type="${1:-}"
 	local session_id="${2:-}"
-	local payload="${3:-{\}}"
+	local payload="${3:-}"
+	[ -z "$payload" ] && payload='{}'
 
 	[[ -z "$event_type" || -z "$session_id" ]] && return 0
 	[[ -z "$_LIBRARIAN_EVENT_JS" || ! -f "$_LIBRARIAN_EVENT_JS" ]] && return 0

package/plugins/tribunal/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "tribunal",
-  "version": "1.1.2",
+  "version": "1.1.3",
   "description": "Multi-agent execution with LLM-as-a-Judge quality gates. An Actor performs work; a jury of typed Judges scores it against a project-overridable rubric; a Meta-Judge reviews the jury for bias; the gate decides accept, retry, or exhaust. Grounded in LLM-as-a-Judge (Zheng et al. 2023) and LLM-as-a-Meta-Judge (Wu et al. 2024). Builds on the Onlooker ecosystem plugin.",
   "author": {
     "name": "Onlooker Community",

package/plugins/tribunal/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [1.1.3](https://github.com/onlooker-community/ecosystem/compare/tribunal-v1.1.2...tribunal-v1.1.3) (2026-08-10)
+
+
+### Bug Fixes
+
+* make the local bats suite tell the truth :mag: ([#135](https://github.com/onlooker-community/ecosystem/issues/135)) ([f0763e0](https://github.com/onlooker-community/ecosystem/commit/f0763e09f3caf2d39c89f28befd12567af0af845))
+
 ## [1.1.2](https://github.com/onlooker-community/ecosystem/compare/tribunal-v1.1.1...tribunal-v1.1.2) (2026-08-02)
 
 

package/plugins/tribunal/scripts/lib/tribunal-aggregate.sh

@@ -25,7 +25,9 @@
 tribunal_aggregate() {
 	local method="${1:-mean}"
 	local verdicts="${2:-[]}"
-	local _rubric="${3:-{}}"  # reserved for true weighted_mean once per-criterion scores are threaded
+	# reserved for true weighted_mean once per-criterion scores are threaded
+	local _rubric="${3:-}"
+	[ -z "$_rubric" ] && _rubric='{}'
 	: "$_rubric"
 
 	local count

package/plugins/tribunal/scripts/lib/tribunal-gate.sh

@@ -23,7 +23,8 @@ tribunal_gate_decide() {
 	local verdicts="${2:-[]}"
 	local aggregated_score="${3:-0}"
 	local score_threshold="${4:-0.75}"
-	local meta="${5:-{}}"
+	local meta="${5:-}"
+	[ -z "$meta" ] && meta='{}'
 	local dissent_score="${6:-0}"
 	local dissent_threshold="${7:-0.25}"
 

package/scripts/lib/prompt-rules.sh

@@ -148,7 +148,12 @@ prompt_rules_pattern_matches() {
 prompt_rules_emit() {
   local session_id="${1:-unknown}"
   local event_type="${2:-}"
-  local payload_json="${3:-{\}}"
+  # Default the payload without inlining braces in the expansion. Both
+  # `${3:-{\}}` and `${3:-{}}` are wrong: the first keeps the backslash on
+  # bash 3.2 (macOS system bash), and the second appends a stray `}` to any
+  # payload that IS supplied, on every bash version.
+  local payload_json="${3:-}"
+  [ -z "$payload_json" ] && payload_json='{}'
   [[ -z "$event_type" ]] && return 1
   ensure_file_exists "$ONLOOKER_EVENTS_LOG" || return 1
 

package/test/bats/archivist-inject.bats

@@ -46,7 +46,7 @@ setup() {
   echo "$output" | jq -e '.hookSpecificOutput.hookEventName == "SessionStart"' >/dev/null
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"use git remote SHA256 as project key"* ]]
+  [[ "$ctx" == *"use git remote SHA256 as project key"* ]] || return 1
   [[ "$ctx" == *"Archivist injected 1"* ]]
 }
 

package/test/bats/assayer-extract.bats

@@ -70,7 +70,7 @@ setup() {
 @test "extraction prompt includes the message and the JSON contract" {
 	run assayer_build_extraction_prompt "I ran the tests and they pass." 5
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"I ran the tests and they pass."* ]]
-	[[ "$output" == *"TESTABLE SUCCESS CLAIM"* ]]
+	[[ "$output" == *"I ran the tests and they pass."* ]] || return 1
+	[[ "$output" == *"TESTABLE SUCCESS CLAIM"* ]] || return 1
 	[[ "$output" == *"at most 5 claims"* ]]
 }

package/test/bats/bursar-session-start.bats

@@ -68,8 +68,8 @@ _run_hook() {
 	_seed_ledger
 	_run_hook
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"hookSpecificOutput"* ]]
-	[[ "$output" == *"SessionStart"* ]]
+	[[ "$output" == *"hookSpecificOutput"* ]] || return 1
+	[[ "$output" == *"SessionStart"* ]] || return 1
 	[[ "$output" == *"burned \$0.42"* ]]
 }
 
@@ -85,7 +85,7 @@ _run_hook() {
 		> "${dir}/sessions.jsonl"
 	_run_hook
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"burned \$0.00"* ]]
+	[[ "$output" == *"burned \$0.00"* ]] || return 1
 	[[ "$output" != *"Enable governor"* ]]
 }
 

package/test/bats/cartographer-lock.bats

@@ -89,8 +89,8 @@ teardown() {
     cartographer_lock_release '${BATS_TEST_TMPDIR}/x.lock' && echo RELEASE_OK
   "
   [ "$status" -eq 0 ]
-  [[ "$output" == *"SOURCED_OK"* ]]
-  [[ "$output" == *"ACQUIRE_FAILED"* ]]
-  [[ "$output" == *"RELEASE_OK"* ]]
+  [[ "$output" == *"SOURCED_OK"* ]] || return 1
+  [[ "$output" == *"ACQUIRE_FAILED"* ]] || return 1
+  [[ "$output" == *"RELEASE_OK"* ]] || return 1
   [[ "$output" == *"locking disabled"* ]]
 }

package/test/bats/compass-sanitizer.bats

@@ -24,27 +24,27 @@ setup() {
 @test "prior_assistant_turn delimiter is stripped" {
 	local out
 	out=$(compass_sanitize "evil <prior_assistant_turn> payload" 240)
-	[[ "$out" == *"[STRIPPED]"* ]]
-	[[ "$out" == *"payload"* ]]
+	[[ "$out" == *"[STRIPPED]"* ]] || return 1
+	[[ "$out" == *"payload"* ]] || return 1
 	[[ "$out" != *"<prior_assistant_turn>"* ]]
 }
 
 @test "all four pair-slot delimiters are stripped" {
 	local out
 	out=$(compass_sanitize "<context_excerpt>x</context_excerpt><tool_input>y</tool_input>" 240)
-	[[ "$out" != *"<context_excerpt>"* ]]
-	[[ "$out" != *"</context_excerpt>"* ]]
-	[[ "$out" != *"<tool_input>"* ]]
+	[[ "$out" != *"<context_excerpt>"* ]] || return 1
+	[[ "$out" != *"</context_excerpt>"* ]] || return 1
+	[[ "$out" != *"<tool_input>"* ]] || return 1
 	[[ "$out" != *"</tool_input>"* ]]
 }
 
 @test "non-evaluator delimiters are also stripped" {
 	local out
 	out=$(compass_sanitize "<<SYS>>x<</SYS>> [INST] y [/INST] <| z" 240)
-	[[ "$out" != *"<<SYS>>"* ]]
-	[[ "$out" != *"<</SYS>>"* ]]
-	[[ "$out" != *"[INST]"* ]]
-	[[ "$out" != *"[/INST]"* ]]
+	[[ "$out" != *"<<SYS>>"* ]] || return 1
+	[[ "$out" != *"<</SYS>>"* ]] || return 1
+	[[ "$out" != *"[INST]"* ]] || return 1
+	[[ "$out" != *"[/INST]"* ]] || return 1
 	[[ "$out" != *"<|"* ]]
 }
 
@@ -53,8 +53,8 @@ setup() {
 	input=$(printf 'a\tb\nc\x00d\x01e')
 	local out
 	out=$(compass_sanitize "$input" 240)
-	[[ "$out" == *"a"* && "$out" == *"b"* && "$out" == *"c"* ]]
-	[[ "$out" == *"d"* && "$out" == *"e"* ]]
+	[[ "$out" == *"a"* && "$out" == *"b"* && "$out" == *"c"* ]] || return 1
+	[[ "$out" == *"d"* && "$out" == *"e"* ]] || return 1
 	[ "${out}" = "$(printf 'a\tb\ncde')" ]
 }
 

package/test/bats/compass-transcript.bats

@@ -60,8 +60,8 @@ setup() {
 	EOF
 	local out
 	out=$(compass_read_prior_turn "$t" 800)
-	[[ "$out" == *"[STRIPPED]"* ]]
-	[[ "$out" != *"<prior_assistant_turn>"* ]]
+	[[ "$out" == *"[STRIPPED]"* ]] || return 1
+	[[ "$out" != *"<prior_assistant_turn>"* ]] || return 1
 
 	# Truncation honors max_chars.
 	local short

package/test/bats/config.bats

@@ -27,7 +27,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PreToolUse[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *tool-sequence-tracker.sh ]]
+  [[ "$hook_cmd" == *tool-sequence-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -41,7 +41,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PreToolUse[1].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *agent-spawn-tracker.sh ]]
+  [[ "$hook_cmd" == *agent-spawn-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -55,7 +55,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PreToolUse[2].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *skill-usage-tracker.sh ]]
+  [[ "$hook_cmd" == *skill-usage-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -78,7 +78,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PostToolUse[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *tool-history-tracker.sh ]]
+  [[ "$hook_cmd" == *tool-history-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -103,9 +103,9 @@ setup_file() {
   turn_cmd=$(jq -r '.hooks.UserPromptSubmit[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
   duration_cmd=$(jq -r '.hooks.UserPromptSubmit[0].hooks[1].command' "${REPO_ROOT}/hooks/hooks.json")
   rule_cmd=$(jq -r '.hooks.UserPromptSubmit[0].hooks[2].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$turn_cmd" == *turn-tracker.sh ]]
-  [[ "$duration_cmd" == *session-duration-tracker.sh ]]
-  [[ "$rule_cmd" == *prompt-rule-injector.sh ]]
+  [[ "$turn_cmd" == *turn-tracker.sh ]] || return 1
+  [[ "$duration_cmd" == *session-duration-tracker.sh ]] || return 1
+  [[ "$rule_cmd" == *prompt-rule-injector.sh ]] || return 1
 
   for hook_cmd in "$turn_cmd" "$duration_cmd" "$rule_cmd"; do
     local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
@@ -121,7 +121,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.SessionStart[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *session-start-tracker.sh ]]
+  [[ "$hook_cmd" == *session-start-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -135,7 +135,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PreCompact[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *pre-compact-tracker.sh ]]
+  [[ "$hook_cmd" == *pre-compact-tracker.sh ]] || return 1
 
   run jq -e '.hooks.PreCompact[0].matcher == "manual" and .hooks.PreCompact[1].matcher == "auto"' \
     "${REPO_ROOT}/hooks/hooks.json"
@@ -153,7 +153,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.PostCompact[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *context-compact-tracker.sh ]]
+  [[ "$hook_cmd" == *context-compact-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -167,7 +167,7 @@ setup_file() {
 
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.SessionEnd[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *session-end-tracker.sh ]]
+  [[ "$hook_cmd" == *session-end-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -178,7 +178,7 @@ setup_file() {
 @test "hooks.json TaskCreated references task-tracker" {
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.TaskCreated[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *task-tracker.sh ]]
+  [[ "$hook_cmd" == *task-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -189,7 +189,7 @@ setup_file() {
 @test "hooks.json TaskCompleted references task-tracker" {
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.TaskCompleted[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *task-tracker.sh ]]
+  [[ "$hook_cmd" == *task-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -200,7 +200,7 @@ setup_file() {
 @test "hooks.json WorktreeCreate references worktree-tracker" {
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.WorktreeCreate[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *worktree-tracker.sh ]]
+  [[ "$hook_cmd" == *worktree-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"
@@ -211,7 +211,7 @@ setup_file() {
 @test "hooks.json WorktreeRemove references worktree-tracker" {
   local hook_cmd
   hook_cmd=$(jq -r '.hooks.WorktreeRemove[0].hooks[0].command' "${REPO_ROOT}/hooks/hooks.json")
-  [[ "$hook_cmd" == *worktree-tracker.sh ]]
+  [[ "$hook_cmd" == *worktree-tracker.sh ]] || return 1
 
   local script_path="${hook_cmd//\$CLAUDE_PLUGIN_ROOT/$REPO_ROOT}"
   script_path="${script_path//\"/}"

package/test/bats/curator-session-start.bats

@@ -35,10 +35,17 @@ setup() {
   # project settings file.
   MEM_DIR="${TEST_HOME}/.claude/projects/test-project/memory"
   mkdir -p "$MEM_DIR" "${PROJECT_REPO}/.claude"
+  # Pin a generous wall-clock budget. The hook checks
+  # cheap_checks.wall_clock_budget_ms before EACH phase and skips the rest
+  # once it trips, so at the shipped default of 500ms a loaded machine can
+  # blow the budget before the later phases run — the orphan check among
+  # them — and those tests fail intermittently while passing in isolation.
+  # These tests assert what the checks detect, not how fast the host is.
   jq -n --arg path "$MEM_DIR" \
     '{
       curator: {
         memory_store_path: $path,
+        cheap_checks: { wall_clock_budget_ms: 600000 },
         date_check: { date_grace_period_days: 7 }
       }
     }' > "${PROJECT_REPO}/.claude/settings.json"
@@ -93,8 +100,8 @@ _write_index() {
   # Surfacer rendered the pointer.
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"Curator: 1 open finding"* ]]
-  [[ "$ctx" == *"date-decayed"* ]]
+  [[ "$ctx" == *"Curator: 1 open finding"* ]] || return 1
+  [[ "$ctx" == *"date-decayed"* ]] || return 1
   [[ "$ctx" == *"/curator review"* ]]
 }
 
@@ -256,7 +263,7 @@ BODY
   [ "$status" -eq 0 ]
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"2 date-decayed"* ]]
+  [[ "$ctx" == *"2 date-decayed"* ]] || return 1
   # The buggy output would have looked like "1 date-decayed, 1 date-decayed".
   [[ "$ctx" != *"1 date-decayed, 1 date-decayed"* ]]
 }

package/test/bats/emit-payload-default.bats

@@ -0,0 +1,52 @@
+#!/usr/bin/env bats
+#
+# Guards the payload-default expansion in every emit helper.
+#
+# Two tempting forms are both wrong, and each fails in a way the other does not:
+#
+#   "${3:-{\}}"  keeps the backslash on bash 3.2 (the bash macOS ships, and the
+#                one `#!/usr/bin/env bash` resolves to there), yielding {\} —
+#                invalid JSON, so the emit silently drops the event. Works on
+#                bash 5.x, which is why CI never caught it.
+#   "${3:-{}}"   appends a stray `}` to any payload that IS supplied, on EVERY
+#                bash version, corrupting the far more common path.
+#
+# The brace-free form below is correct on both. This test runs the real
+# expansion under the real interpreter rather than asserting on source text.
+
+setup() {
+	source "${BATS_TEST_DIRNAME}/../helpers/setup.bash"
+	setup_test_env
+}
+
+_emit_libs() {
+	printf '%s\n' \
+		"scripts/lib/prompt-rules.sh" \
+		"plugins/curator/scripts/lib/curator-emit.sh" \
+		"plugins/historian/scripts/lib/historian-emit.sh" \
+		"plugins/librarian/scripts/lib/librarian-emit.sh" \
+		"plugins/librarian/scripts/lib/librarian-cli.sh" \
+		"plugins/tribunal/scripts/lib/tribunal-gate.sh"
+}
+
+@test "no emit helper inlines braces in a parameter-expansion default" {
+	# Code only — a comment explaining the hazard must not trip its own guard.
+	local hits
+	hits=$(cd "$REPO_ROOT" && grep -rn ':-{' --include="*.sh" scripts/ plugins/ \
+		| grep -vE '^[^:]+:[0-9]+:[[:space:]]*#' || true)
+	[ -z "$hits" ]
+}
+
+@test "the brace-free default yields {} when no payload is passed" {
+	local probe="${BATS_TEST_TMPDIR}/probe.sh"
+	printf '%s\n' 'f() { local p="${3:-}"; [ -z "$p" ] && p='"'"'{}'"'"'; printf "%s" "$p"; }' 'f a b' >"$probe"
+	run bash "$probe"
+	[ "$status" -eq 0 ] && [ "$output" = '{}' ]
+}
+
+@test "the brace-free default passes a supplied payload through unchanged" {
+	local probe="${BATS_TEST_TMPDIR}/probe.sh"
+	printf '%s\n' 'f() { local p="${3:-}"; [ -z "$p" ] && p='"'"'{}'"'"'; printf "%s" "$p"; }' 'f a b '"'"'{"a":{"b":1}}'"'"'' >"$probe"
+	run bash "$probe"
+	[ "$status" -eq 0 ] && [ "$output" = '{"a":{"b":1}}' ]
+}

package/test/bats/governor-ledger.bats

@@ -91,7 +91,7 @@ _make_completion() {
 @test "governor_ledger_path returns a .jsonl path under ONLOOKER_DIR" {
 	local p
 	p=$(governor_ledger_path "test-session")
-	[[ "$p" == *".jsonl" ]]
+	[[ "$p" == *".jsonl" ]] || return 1
 	[[ "$p" == *"test-session"* ]]
 }
 

package/test/bats/historian-prompt-submit.bats

@@ -157,7 +157,7 @@ _index_session() {
 
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"Historian: a past chunk looks similar"* ]]
+  [[ "$ctx" == *"Historian: a past chunk looks similar"* ]] || return 1
   [[ "$ctx" == *"redash"* ]]
 
   grep '"event_type":"historian.retrieval.surfaced"' "$ONLOOKER_EVENTS_LOG" \

package/test/bats/inspector-post-write-hook.bats

@@ -100,7 +100,7 @@ _run_hook() {
 	echo '{"inspector":{"show_clean_runs":true,"checks":{".ts":[{"name":"clean","kind":"lint","argv":["true"]}]}}}' | _settings
 	run _run_hook "$(_input)"
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"inspector: src/sample.ts"* ]]
+	[[ "$output" == *"inspector: src/sample.ts"* ]] || return 1
 	[[ "$output" == *"✓ clean"* ]]
 }
 
@@ -112,9 +112,9 @@ _run_hook() {
 EOF
 	run _run_hook "$(_input)"
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"inspector: src/sample.ts"* ]]
-	[[ "$output" == *"✗ broken"* ]]
-	[[ "$output" == *"src/sample.ts:1:1 - Bad"* ]]
+	[[ "$output" == *"inspector: src/sample.ts"* ]] || return 1
+	[[ "$output" == *"✗ broken"* ]] || return 1
+	[[ "$output" == *"src/sample.ts:1:1 - Bad"* ]] || return 1
 	[ "$(_event_count inspector.check.failed)" = "1" ]
 	[ "$(jq -r 'select(.event_type=="inspector.check.failed").payload.exit_code' "$ONLOOKER_EVENTS_LOG")" = "2" ]
 }

package/test/bats/librarian-cli.bats

@@ -124,13 +124,13 @@ _seed_proposal() {
   run librarian_cli_list "$PROJECT_REPO"
   [ "$status" -eq 0 ]
   # Header counts only pending entries (2 of 3).
-  [[ "$output" == *"2 pending proposal"* ]]
+  [[ "$output" == *"2 pending proposal"* ]] || return 1
   # Both pending titles surface.
-  [[ "$output" == *"Prefer functional patterns"* ]]
-  [[ "$output" == *"Auth rewrite is compliance"* ]]
+  [[ "$output" == *"Prefer functional patterns"* ]] || return 1
+  [[ "$output" == *"Auth rewrite is compliance"* ]] || return 1
   # Pending rows include full IDs that can be used with show/accept/reject/defer.
-  [[ "$output" == *"01LISTPENDINGA000000000000"* ]]
-  [[ "$output" == *"01LISTPENDINGB000000000000"* ]]
+  [[ "$output" == *"01LISTPENDINGA000000000000"* ]] || return 1
+  [[ "$output" == *"01LISTPENDINGB000000000000"* ]] || return 1
   # Accepted entry does NOT appear in the list output.
   [[ "$output" != *"Already accepted"* ]]
 }
@@ -142,10 +142,10 @@ _seed_proposal() {
 
   run librarian_cli_show "01SHOWPROPOSAL0000000000000" "$PROJECT_REPO"
   [ "$status" -eq 0 ]
-  [[ "$output" == *"01SHOWPROPOSAL0000000000000"* ]]
-  [[ "$output" == *"type:                 feedback"* ]]
-  [[ "$output" == *"filename:             feedback_x.md"* ]]
-  [[ "$output" == *"classifier_confidence: 0.82"* ]]
+  [[ "$output" == *"01SHOWPROPOSAL0000000000000"* ]] || return 1
+  [[ "$output" == *"type:                 feedback"* ]] || return 1
+  [[ "$output" == *"filename:             feedback_x.md"* ]] || return 1
+  [[ "$output" == *"classifier_confidence: 0.82"* ]] || return 1
   [[ "$output" == *"Body of the memory."* ]]
 }
 
@@ -165,7 +165,7 @@ _seed_proposal() {
 
   run librarian_cli_accept "01ACCEPTPROPOSAL000000000000" "$PROJECT_REPO"
   [ "$status" -eq 0 ]
-  [[ "$output" == *"Accepted."* ]]
+  [[ "$output" == *"Accepted."* ]] || return 1
 
   local out_file="${MEM_DIR}/feedback_functional.md"
   [ -f "$out_file" ]
@@ -220,7 +220,7 @@ _seed_proposal() {
 
   run librarian_cli_accept "01ACCEPTUNSAFE00000000000000" "$PROJECT_REPO"
   [ "$status" -eq 1 ]
-  [[ "$output" == *"Failed to write memory file."* ]]
+  [[ "$output" == *"Failed to write memory file."* ]] || return 1
 
   # No memory file written anywhere under MEM_DIR or its parent.
   [ ! -f "${MEM_DIR}/../escape.md" ]
@@ -238,8 +238,8 @@ _seed_proposal() {
 
   run librarian_cli_reject "01REJECTPROPOSAL00000000000" "stale guidance" "$PROJECT_REPO"
   [ "$status" -eq 0 ]
-  [[ "$output" == *"Rejected"* ]]
-  [[ "$output" == *"stale guidance"* ]]
+  [[ "$output" == *"Rejected"* ]] || return 1
+  [[ "$output" == *"stale guidance"* ]] || return 1
 
   # Proposal status now rejected with the reason captured.
   local proposal_path="${LIBRARIAN_DIR}/proposals/01REJECTPROPOSAL00000000000.json"
@@ -268,7 +268,7 @@ _seed_proposal() {
 
   run librarian_cli_defer "01DEFERPROPOSAL000000000000" "$PROJECT_REPO"
   [ "$status" -eq 0 ]
-  [[ "$output" == *"Deferred"* ]]
+  [[ "$output" == *"Deferred"* ]] || return 1
 
   local proposal_path="${LIBRARIAN_DIR}/proposals/01DEFERPROPOSAL000000000000.json"
   jq -e '.status == "pending"' "$proposal_path" >/dev/null
@@ -285,8 +285,8 @@ _seed_proposal() {
 
   run librarian_cli_status "$PROJECT_REPO"
   [ "$status" -eq 0 ]
-  [[ "$output" == *"pending: 1"* ]]
-  [[ "$output" == *"accepted: 1"* ]]
+  [[ "$output" == *"pending: 1"* ]] || return 1
+  [[ "$output" == *"accepted: 1"* ]] || return 1
   [[ "$output" == *"rejected: 1"* ]]
 }
 

package/test/bats/librarian-session-start.bats

@@ -83,7 +83,7 @@ _seed_proposal() {
   [ "$status" -eq 0 ]
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"Librarian has 3 pending memory promotion proposals"* ]]
+  [[ "$ctx" == *"Librarian has 3 pending memory promotion proposals"* ]] || return 1
   [[ "$ctx" == *"/librarian review"* ]]
 }
 
@@ -94,7 +94,7 @@ _seed_proposal() {
   [ "$status" -eq 0 ]
   local ctx
   ctx=$(echo "$output" | jq -r '.hookSpecificOutput.additionalContext')
-  [[ "$ctx" == *"Librarian has 1 pending memory promotion proposal"* ]]
+  [[ "$ctx" == *"Librarian has 1 pending memory promotion proposal"* ]] || return 1
   [[ "$ctx" != *"proposals."* ]]
 }
 

package/test/bats/lineage-config.bats

@@ -43,6 +43,6 @@ setup() {
 	lineage_config_load ""
 	run lineage_config_ignore_globs
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"node_modules"* ]]
+	[[ "$output" == *"node_modules"* ]] || return 1
 	[[ "$output" == *".lock"* ]]
 }

package/test/bats/lineage-redact.bats

@@ -21,7 +21,7 @@ setup() {
 	local tok out
 	tok="sk-ant-$(printf 'a%.0s' $(seq 1 28))"
 	out=$(printf '%s' "key=\"${tok}\"" | lineage_redact 4000 true)
-	[[ "$out" == *"[REDACTED:secret]"* ]]
+	[[ "$out" == *"[REDACTED:secret]"* ]] || return 1
 	[[ "$out" != *"$tok"* ]]
 }
 
@@ -29,7 +29,7 @@ setup() {
 	local tok out
 	tok="ghp_$(printf '0%.0s' $(seq 1 36))"
 	out=$(printf '%s' "$tok" | lineage_redact 4000 true)
-	[[ "$out" == *"[REDACTED:secret]"* ]]
+	[[ "$out" == *"[REDACTED:secret]"* ]] || return 1
 	[[ "$out" != *"$tok"* ]]
 }
 
@@ -37,8 +37,8 @@ setup() {
 	local val out
 	val=$(printf 'x%.0s' $(seq 1 24))
 	out=$(printf '%s' "API_TOKEN=${val}" | lineage_redact 4000 true)
-	[[ "$out" == *"API_TOKEN="* ]]
-	[[ "$out" == *"[REDACTED:secret]"* ]]
+	[[ "$out" == *"API_TOKEN="* ]] || return 1
+	[[ "$out" == *"[REDACTED:secret]"* ]] || return 1
 	[[ "$out" != *"$val"* ]]
 }
 
@@ -52,7 +52,7 @@ setup() {
 @test "content longer than the cap is truncated with a marker" {
 	local out
 	out=$(printf '%s' "abcdefghij" | lineage_redact 4 true)
-	[[ "$out" == "abcd"* ]]
+	[[ "$out" == "abcd"* ]] || return 1
 	[[ "$out" == *"truncated 6 chars"* ]]
 }
 

package/test/bats/session-tracker.bats

@@ -90,7 +90,7 @@ seed_start_ms_ago() {
 @test "session_tracker_now_ms returns 13-digit epoch milliseconds" {
   local ms
   ms=$(session_tracker_now_ms)
-  [[ "$ms" =~ ^[0-9]+$ ]]
+  [[ "$ms" =~ ^[0-9]+$ ]] || return 1
   [ "${#ms}" -eq 13 ]
 }
 
@@ -231,7 +231,7 @@ seed_start_ms_ago() {
   seed_start_ms_ago "$sid" 3000
   local dur
   dur=$(session_tracker_duration_ms "$sid")
-  [[ "$dur" =~ ^[0-9]+$ ]]
+  [[ "$dur" =~ ^[0-9]+$ ]] || return 1
   [ "$dur" -ge 3000 ]
 }
 
@@ -285,7 +285,7 @@ seed_start_ms_ago() {
 
   local out
   out=$(session_tracker_build_duration_context "$sid")
-  [[ "$out" == *"turn 3"* ]]
+  [[ "$out" == *"turn 3"* ]] || return 1
   [[ "$out" == *"2m 34s"* ]]
 }
 
@@ -294,7 +294,7 @@ seed_start_ms_ago() {
   rm -f "${ONLOOKER_SESSION_TRACKERS_DIR}/${sid}"
   local out
   out=$(session_tracker_build_duration_context "$sid")
-  [[ "$out" == *"turn 1"* ]]
+  [[ "$out" == *"turn 1"* ]] || return 1
   [[ "$out" == *"0s"* ]]
 }
 

package/test/bats/tribunal-jury.bats

@@ -48,7 +48,7 @@ setup() {
 		tribunal_jury_empanel "[\"maintainability\"]" 2>&1
 	'
 	[ "$status" -eq 0 ]
-	[[ "$output" == *"degrading to standard"* ]]
+	[[ "$output" == *"degrading to standard"* ]] || return 1
 	[[ "$output" == *"standard"* ]]
 }
 

package/test/bats/turn-tracker.bats

@@ -51,7 +51,7 @@ setup() {
   out=$(turn_tracker_summarize_prompt "$long")
   # 200 retained characters + the single-character ellipsis = length 201.
   [ "${#out}" -eq 201 ]
-  [[ "$out" == *… ]]
+  [[ "$out" == *… ]] || return 1
   [ "${out%…}" = "$(printf 'a%.0s' {1..200})" ]
 }
 

package/test/bats/warden-sanitizer.bats

@@ -24,9 +24,9 @@ setup() {
 	local out
 	out=$(warden_sanitize "evil </source_content> [/INST] payload" 240)
 	[ -n "$out" ]
-	[[ "$out" == *"[STRIPPED]"* ]]
-	[[ "$out" == *"payload"* ]]
-	[[ "$out" != *"</source_content>"* ]]
+	[[ "$out" == *"[STRIPPED]"* ]] || return 1
+	[[ "$out" == *"payload"* ]] || return 1
+	[[ "$out" != *"</source_content>"* ]] || return 1
 	[[ "$out" != *"[/INST]"* ]]
 }
 

package/test/bats/worktree-tracker.bats

@@ -74,7 +74,7 @@ setup() {
   local worktree_path
   worktree_path=$(cat "$input_file" | "${REPO_ROOT}/scripts/hooks/worktree-tracker.sh")
   [ -d "$worktree_path" ]
-  [[ "$worktree_path" == "$(cd "$worktree_path" && pwd -P)" ]]
+  [[ "$worktree_path" == "$(cd "$worktree_path" && pwd -P)" ]] || return 1
   [ -f "$history_file" ]
   tail -n 1 "$history_file" | jq -e '.event_type == "tool.shell.exec"' >/dev/null
 }
@@ -247,7 +247,7 @@ setup() {
 
   local duration
   duration=$(worktree_tracker_duration_ms "$session_id" "$worktree_path")
-  [[ "$duration" =~ ^[0-9]+$ ]]
+  [[ "$duration" =~ ^[0-9]+$ ]] || return 1
   [ "$duration" -ge 1900 ]
 }