@gobing-ai/spur 0.4.8 → 0.4.10

package/plugins/sp/scripts/hooks/careful-guard.ts

@@ -0,0 +1,94 @@
+#!/usr/bin/env bun
+
+/**
+ * careful-guard — PreToolUse guard for destructive shell commands (task 0215, R3).
+ *
+ * Warn (permission `ask`) before a `Bash` tool call runs a destructive command — `rm -rf`
+ * (any flag spelling, including the POSIX uppercase `-R`), `DROP TABLE`/`DROP DATABASE`/
+ * `TRUNCATE`, `git push --force`/`-f`/`+refspec`, `git reset --hard`, `git checkout .`/
+ * `git restore .`, `kubectl delete`, `docker system prune`. The operator can confirm to
+ * proceed. Pure pattern-match + decision, no domain logic (mirrors `task-write-guard`).
+ *
+ * Git global options between the executable and the subcommand are recognized — any run of them,
+ * not a fixed list: value-less flags (`--no-pager`, `--bare`, `-p`, …), glued long values
+ * (`--git-dir=<path>`), and value-taking options in separated form (`-C <path>`,
+ * `-c <name>=<value>`, `--git-dir|--work-tree|--namespace|--config-env|--attr-source <value>`).
+ * So `git -C . push --force` and `git --no-pager push --force` classify exactly like
+ * `git push --force`; the lexical pass only removes that option span and never executes git or
+ * expands a substitution. The run cannot cross a command boundary (`;`, `&`, `|`, newline).
+ *
+ * **Safe exceptions:** `rm -rf` of well-known build/dependency caches (`node_modules`, `dist`,
+ * `.next`, `coverage`, `build`, `.turbo`, `.cache`) passes without a warning — deleting a rebuildable
+ * cache is routine, not dangerous.
+ *
+ * **Output contract** (superskill ADR-039 `hook run`): allow prints **nothing** (exit 0). Ask prints
+ * Claude-native PreToolUse JSON with the reason in `permissionDecisionReason` (the confirm prompt) and
+ * `systemMessage` (Claude Code's user-facing warning, which the host-neutral `{decision,reason}` form
+ * cannot carry). Hosts that cannot confirm (Codex, omp, generic) receive a deny with the reason:
+ * superskill fails an unconfirmable ask closed. Pi confirms through its native host adapter.
+ *
+ * **Fail-open contract:** every error path — unparseable payload, non-Bash tool, empty command —
+ * allows (prints nothing). A broken guard must never wedge an agent tool call.
+ *
+ * **Escape hatch:** `SPUR_CAREFUL=off` short-circuits to allow.
+ *
+ * Classification itself lives in `lib/destructive-policy.ts` — one policy, every platform
+ * adapter runs this hook through superskill). This file owns the hook payload/decision shape only.
+ * Registered in `hooks/hooks.json` as `PreToolUse` / matcher `Bash` (task 0215 R3).
+ */
+
+import { classifyCommand } from '../../lib/destructive-policy';
+import { getEnvVar } from '../../lib/env';
+
+interface ToolPayload {
+    tool_name?: string;
+    tool_input?: { command?: string };
+}
+
+/** Allow: print nothing and exit 0 (no objection; the host's own permission rules still apply). */
+function allow(): never {
+    process.exit(0);
+}
+
+/** Ask the operator to confirm, with the reason on the prompt and as a user-facing warning. */
+function ask(reason: string): never {
+    process.stdout.write(
+        JSON.stringify({
+            hookSpecificOutput: {
+                hookEventName: 'PreToolUse',
+                permissionDecision: 'ask',
+                permissionDecisionReason: reason,
+            },
+            systemMessage: reason,
+        }),
+    );
+    process.exit(0);
+}
+
+async function main(): Promise<void> {
+    if (getEnvVar('SPUR_CAREFUL') === 'off') allow();
+
+    const stdinText = await Bun.stdin.text();
+    let payload: ToolPayload;
+    try {
+        payload = JSON.parse(stdinText) as ToolPayload;
+    } catch {
+        allow(); // unparseable payload — fail open
+    }
+
+    if (payload.tool_name !== 'Bash') allow();
+
+    const command = payload.tool_input?.command ?? '';
+    if (command.trim() === '') allow();
+
+    const hit = classifyCommand(command);
+    if (hit !== null) {
+        ask(
+            `This command looks like ${hit} — a destructive operation. Confirm you intend to run it. ` +
+                'Set SPUR_CAREFUL=off to disable this guard.',
+        );
+    }
+    allow();
+}
+
+if (import.meta.main) void main();

package/plugins/sp/{hooks → scripts/hooks}/context-post-tool.ts

@@ -10,8 +10,8 @@
  */
 
 import { join } from 'node:path';
-import { recordToolUseEvent, type ToolPayload } from '../lib/context-post-tool';
-import { getEnvVar } from '../lib/env';
+import { recordToolUseEvent, type ToolPayload } from '../../lib/context-post-tool';
+import { getEnvVar } from '../../lib/env';
 
 if (import.meta.main) {
     void (async () => {

package/plugins/sp/{hooks → scripts/hooks}/context-session-start.ts

@@ -7,7 +7,7 @@
  * (legacy policy) and every error path exits 0 with no output.
  */
 
-import { recordSessionStartFromPayloadText } from '../lib/context-session-start';
+import { recordSessionStartFromPayloadText } from '../../lib/context-session-start';
 
 if (import.meta.main) {
     try {

package/plugins/sp/{hooks → scripts/hooks}/context-session-stop.ts

@@ -7,7 +7,7 @@
  * `agent.run`'s shutdown stays a no-op because the owner gate lives in the core, not here.
  */
 
-import { recordSessionEndFromPayloadText } from '../lib/context-session-stop';
+import { recordSessionEndFromPayloadText } from '../../lib/context-session-stop';
 
 if (import.meta.main) {
     try {

package/plugins/sp/{hooks → scripts/hooks}/fleet-guest-stop.ts

@@ -15,9 +15,11 @@
  * - **Bounded.** The pending check is `spur agent wait --inbox <id> --timeout 2000`, which also
  *   heartbeats the guest lease (R5), and the reason names at most the first few messages.
  *
- * Block contract (both hosts that support it): exit 0 with JSON on stdout —
- * `{"decision":"block","reason":"<non-empty>"}`. Claude Code uses the reason as the stop reason;
- * Codex turns it into the next user prompt. Hosts that drop the Stop event (`codex exec` on
+ * Block contract: exit 0 with the host-neutral superskill result on stdout (ADR-039) —
+ * `{"decision":"block","reason":"<non-empty>"}`. `superskill hook run` renders it per host: Claude
+ * Code and Codex receive that same Stop JSON (Claude uses the reason as the stop reason; Codex turns
+ * it into the next user prompt), omp/Pi get exit 2 + the reason on stderr. No pending work prints
+ * nothing. Hosts that drop the Stop event (`codex exec` on
  * 0.160.0 — see `docs/plans/2026-10-05-codex-guest-spike.md`) simply never call it.
  *
  * Self-contained by design: hooks are bundled standalone at install time, so this file imports
@@ -27,7 +29,7 @@
 import { spawnSync } from 'node:child_process';
 import { existsSync, readdirSync, readFileSync } from 'node:fs';
 import { join } from 'node:path';
-import { getEnvVar } from '../lib/env';
+import { getEnvVar } from '../../lib/env';
 
 /** Budget for the pending-work probe; the hook must never noticeably slow a turn. */
 export const GUEST_STOP_PROBE_MS = 2000;

package/plugins/sp/{hooks → scripts/hooks}/task-write-guard.ts

@@ -7,21 +7,28 @@
  * hand. Pure delegation: ownership is decided by `spur task resolve --strict --json`'s exit code
  * alone (exit 0 → owned, non-zero → unowned).
  *
+ * **Output contract** (superskill ADR-039 `hook run`): allow prints **nothing** (exit 0) — a printed
+ * Claude `permissionDecision:"allow"` would auto-approve the call and bypass the user's permission
+ * rules. Deny prints Claude-native PreToolUse JSON carrying the reason twice: `permissionDecisionReason`
+ * (what Claude Code feeds back to the model, and what `hook run` turns into exit 2 + stderr on
+ * Codex/omp/generic hosts) and `systemMessage` (Claude Code's user-facing warning — the one field the
+ * host-neutral `{decision,reason}` form cannot carry, which is why this guard stays Claude-native).
+ *
  * **Fail-open contract** (inherited from the pre-N3 shim and from superskill's runner): every error
  * path — unparseable payload, non-Write/Edit tool, empty path, missing `spur` binary, spawn
- * failure, timeout, `spur task resolve` non-zero exit, non-JSON stdout — emits an `allow`
- * decision. A broken guard must never wedge an agent tool call.
+ * failure, timeout, `spur task resolve` non-zero exit, non-JSON stdout — allows (prints nothing).
+ * A broken guard must never wedge an agent tool call.
  *
  * **Escape hatch:** `SPUR_WRITE_GUARD=off` short-circuits to allow.
  *
- * Self-contained by design (Wave E, task 0181, R3). Installed hook configs still use the portable
- * `superskill hook run sp task-write-guard` entrypoint from task 0151, while this source script keeps
- * the real guard decisions versioned and covered by this repo's test gate.
+ * Self-contained by design (Wave E, task 0181, R3): `node:*` builtins and relative imports only.
+ * Installed hook configs invoke `superskill hook run sp task-write-guard`, which resolves this file
+ * as `scripts/hooks/task-write-guard.ts` under the native plugin root or the staged scripts root.
  */
 
 import { spawnSync } from 'node:child_process';
-import { getEnvVar } from '../lib/env';
-import { couldBeTaskFile } from '../lib/task-file-policy';
+import { getEnvVar } from '../../lib/env';
+import { couldBeTaskFile } from '../../lib/task-file-policy';
 
 interface ToolPayload {
     tool_name?: string;
@@ -30,12 +37,23 @@ interface ToolPayload {
 
 type TaskOwnership = 'owned' | 'unowned' | 'unknown';
 
-function preToolUseDecision(decision: 'allow' | 'deny', reason?: string): never {
-    const out: Record<string, unknown> = {
-        hookSpecificOutput: { hookEventName: 'PreToolUse', permissionDecision: decision },
-    };
-    if (reason !== undefined) out.systemMessage = reason;
-    process.stdout.write(JSON.stringify(out));
+/** Allow: print nothing and exit 0 (no objection; the host's own permission rules still apply). */
+function allow(): never {
+    process.exit(0);
+}
+
+/** Deny with a reason the model (`permissionDecisionReason`) and the user (`systemMessage`) both see. */
+function deny(reason: string): never {
+    process.stdout.write(
+        JSON.stringify({
+            hookSpecificOutput: {
+                hookEventName: 'PreToolUse',
+                permissionDecision: 'deny',
+                permissionDecisionReason: reason,
+            },
+            systemMessage: reason,
+        }),
+    );
     process.exit(0);
 }
 
@@ -55,35 +73,34 @@ function resolveSpurTaskOwnership(filePath: string, cwd: string): TaskOwnership
 }
 
 async function main(): Promise<void> {
-    if (getEnvVar('SPUR_WRITE_GUARD') === 'off') preToolUseDecision('allow');
+    if (getEnvVar('SPUR_WRITE_GUARD') === 'off') allow();
 
     const stdinText = await Bun.stdin.text();
     let payload: ToolPayload;
     try {
         payload = JSON.parse(stdinText) as ToolPayload;
     } catch {
-        preToolUseDecision('allow'); // unparseable payload — fail open
+        allow(); // unparseable payload — fail open
     }
 
     const toolName = payload.tool_name ?? '';
-    if (toolName !== 'Write' && toolName !== 'Edit') preToolUseDecision('allow');
+    if (toolName !== 'Write' && toolName !== 'Edit') allow();
 
     const filePath = payload.tool_input?.file_path ?? '';
-    if (filePath === '') preToolUseDecision('allow');
+    if (filePath === '') allow();
     // Skip the `spur task resolve` subprocess (~120ms) for paths whose basename
     // cannot name a task file — that is every ordinary source edit.
-    if (!couldBeTaskFile(filePath)) preToolUseDecision('allow');
+    if (!couldBeTaskFile(filePath)) allow();
 
     const ownership = resolveSpurTaskOwnership(filePath, getEnvVar('CLAUDE_PROJECT_DIR') ?? process.cwd());
     if (ownership === 'owned') {
-        preToolUseDecision(
-            'deny',
+        deny(
             `${filePath} is a task file owned by the spur corpus. Edit it through the spur CLI ` +
                 '(e.g. `spur task update <wbs> --section <name> --from-file <file>`), not a raw ' +
                 'Write/Edit. Set SPUR_WRITE_GUARD=off to bypass.',
         );
     }
-    preToolUseDecision('allow');
+    allow();
 }
 
 void main();

package/plugins/sp/skills/brainstorm/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: brainstorm
-description: "Structured ideation: generate solution options with trade-offs and confidence scoring, then delegate to research/task-creation skills. Triggers: \"brainstorm ideas\", \"explore solutions\", \"consider options\", \"research approaches\", \"what are my options\", \"how should I approach X\"."
+description: "Structured ideation: generate solution options with trade-offs and source citations, then delegate to research/task-creation skills. Triggers: \"brainstorm ideas\", \"explore solutions\", \"consider options\", \"research approaches\", \"what are my options\", \"how should I approach X\"."
 license: Apache-2.0
 version: 1.0.0
 created_at: 2026-03-25
@@ -30,7 +30,7 @@ see_also:
 
 # sp:brainstorm — Structured Ideation Workflow
 
-Generate solution options with trade-offs, recommendations, and confidence scoring. Delegates research to specialized skills.
+Generate solution options with trade-offs, recommendations, and source citations. Delegates research to specialized skills.
 
 **Key distinction:**
 - **`sp:brainstorm`** = Ideation: generate approaches with trade-offs
@@ -40,7 +40,7 @@ Generate solution options with trade-offs, recommendations, and confidence scori
 
 ## Overview
 
-The `sp:brainstorm` skill generates multiple solution approaches with explicit trade-offs, confidence scoring, and source citations. It follows a structured 3-phase workflow: Input parsing, Ideation with research delegation, and structured Output. Unlike pure research or bare task creation, brainstorm focuses on ideation—generating and comparing options before committing to a solution path.
+The `sp:brainstorm` skill generates multiple solution approaches with explicit trade-offs and source citations. It follows a structured 3-phase workflow: Input parsing, Ideation with research delegation, and structured Output. Unlike pure research or bare task creation, brainstorm focuses on ideation—generating and comparing options before committing to a solution path.
 
 ## Quick Start
 
@@ -101,14 +101,15 @@ See the [inline-default contract](../spur-dev/references/cross-cutting.md#inline
 
 ### 3. Generate 2-3 Approaches
 
-Always generate multiple options, each with description, trade-offs, confidence, and sources — full
+Always generate multiple options, each with description, trade-offs, and sources — full
 per-approach template: [references/workflows.md](references/workflows.md#approach-generation).
 
-### 4. Confidence Scoring
+### 4. Source Citations
 
-Every approach and every external claim carries a HIGH/MEDIUM/LOW confidence score plus a dated
-source citation — table, thresholds, and citation format:
-[references/workflows.md](references/workflows.md#confidence-scoring).
+Every external claim carries a dated source citation — format:
+[references/workflows.md](references/workflows.md#source-citation-format). No HIGH/MEDIUM/LOW
+confidence score (ADR-138): a self-assigned score is as easy to write as the claim; the citation and
+the trade-offs are what the operator compares.
 
 ### 5. Task Delegation
 
@@ -224,13 +225,13 @@ still pauses unless the operator has encoded prior approval in the workflow vars
 | "I understand the request — skip clarifying." | Ideating on a misread wastes the whole session. Validate input clarity and the two input modes before generating options. |
 | "More options are always better." | Beyond ~3, options dilute focus and stall the decision. Cap at three and delegate deeper research instead. |
 | "I'll design the architecture while I brainstorm." | Brainstorm generates and scores approaches; it does not lock the design. Route a chosen approach through the Design Approval Gate. |
-| "Confidence scores are subjective — skip them." | An unscored option hides its risk. Cite sources and assign confidence so the operator compares on evidence, not vibes. |
+| "Sources are obvious — skip them." | An uncited option hides its risk. Cite sources and name the trade-offs so the operator compares on evidence, not vibes. |
 
 ## Red Flags
 
 - Presenting a single approach with no alternatives or trade-offs.
 - Starting ideation before the input is clarified.
-- More than three approaches, or approaches with no confidence score / source.
+- More than three approaches, or approaches with no source.
 - Skipping the task-delegation offer after the operator confirms a direction.
 - Treating a brainstorm output as an approved design (bypassing the approval gate).
 
@@ -293,7 +294,7 @@ codebase-first rule, stop conditions, depth question budgets, smart positional d
 The skill uses this tree to:
 1. **Constrain the option space** — each approach must be compatible with resolved decisions
 2. **Generate decision-trace annotations** — each approach lists which decisions it depends on
-3. **Calibrate confidence** — decisions resolved from codebase evidence get higher confidence
+3. **Cite the evidence** — decisions resolved from codebase evidence cite the `path:line` they rest on
 
 When `--skip-discovery` is used (no `--context`), the operation falls back to the standard
 3-phase workflow with its own lightweight clarification step.

package/plugins/sp/skills/brainstorm/agents/openai.yaml

@@ -1,6 +1,6 @@
 name: brainstorm
 description: "Structured ideation workflow for generating solution options with
-  trade-offs, confidence scoring, and delegation to research and task creation
+  trade-offs, source citations, and delegation to research and task creation
   skills. Triggers: brainstorm ideas, explore solutions, consider options,
   research approaches, multiple solution options with trade-offs."
 version: 1.0.0

package/plugins/sp/skills/brainstorm/examples/ideation-example.md

@@ -80,7 +80,6 @@ async function authMiddleware(c: Context, next: Next) {
 }
 ```
 
-**Confidence:** HIGH
 **Sources:**
 - [jose Library Docs](https://github.com/panva/jose) | Verified: 2026-03-25
 - [Hono Middleware Guide](https://hono.dev/docs/middleware/builtin/bearer-auth) | Verified: 2026-03-25
@@ -127,7 +126,6 @@ async function handleLogin(username: string, password: string) {
 }
 ```
 
-**Confidence:** MEDIUM
 **Sources:**
 - [Auth0 Node.js SDK v5](https://auth0.github.io/node-auth0/) | Verified: 2026-03-25
 - [Firebase Admin Setup](https://firebase.google.com/docs/admin/setup) | Verified: 2026-03-15
@@ -172,7 +170,6 @@ async function apiKeyMiddleware(c: Context, next: Next) {
 }
 ```
 
-**Confidence:** MEDIUM
 **Sources:**
 - [Hono Bearer Auth Middleware](https://hono.dev/docs/middleware/builtin/bearer-auth) | Verified: 2026-03-25
 
@@ -206,4 +203,3 @@ async function apiKeyMiddleware(c: Context, next: Next) {
 
 **Generated by:** sp:brainstorm
 **Research delegation:** sp:source-driven-development, spur agent run
-**Confidence:** HIGH (official jose/Hono documentation, 2026)

package/plugins/sp/skills/brainstorm/metadata.openclaw

@@ -1,6 +1,6 @@
 name: brainstorm
 description: "Structured ideation workflow for generating solution options with
-  trade-offs, confidence scoring, and delegation to research and task creation
+  trade-offs, source citations, and delegation to research and task creation
   skills. Triggers: brainstorm ideas, explore solutions, consider options,
   research approaches, multiple solution options with trade-offs."
 version: 1.0.0

package/plugins/sp/skills/brainstorm/references/workflows.md

@@ -85,7 +85,7 @@ DO NOT implement research directly. Delegate to specialized skills:
 
 2. Delegate research + synthesis inline by default
    - Run in the current session; escalate to `spur agent run` only on a subprocess trigger
-   - Request confidence scoring
+   - Request source citations
 ```
 
 ### Approach Generation
@@ -108,7 +108,6 @@ DO NOT implement research directly. Delegate to specialized skills:
 - Key technical considerations
 - Dependencies or prerequisites
 
-**Confidence:** HIGH/MEDIUM/LOW
 **Sources:** [Citations with dates]
 ```
 
@@ -120,14 +119,6 @@ DO NOT implement research directly. Delegate to specialized skills:
 - [Title](URL) | **Verified**: YYYY-MM-DD
 ```
 
-### Confidence Scoring
-
-| Level | Range | Criteria |
-|-------|-------|----------|
-| **HIGH** | >90% | Direct quote from official docs (2025+) |
-| **MEDIUM** | 70-90% | Synthesized from multiple sources |
-| **LOW** | <70% | Uncertain, flag for review |
-
 ---
 
 ## Phase 3: Structured Output
@@ -144,7 +135,7 @@ DO NOT implement research directly. Delegate to specialized skills:
 2. **Approaches** (200-300 words each)
    - 2-3 options with trade-offs
    - Implementation considerations
-   - Confidence levels
+   - Sources
 
 3. **Recommendations** (100-150 words)
    - Recommended approach with reasoning
@@ -197,7 +188,6 @@ run_id: <run id, omit when none>
 **Implementation Notes:**
 - [Technical considerations]
 
-**Confidence:** HIGH/MEDIUM/LOW
 **Sources:** [Citations]
 
 ### Approach 2: [Name]
@@ -255,7 +245,7 @@ Bash: spur task batch-create --file decomposition.json   # bare JSON array (see
 | File not found | Provide clear error with path |
 | Invalid YAML | Report parsing error |
 | Missing fields | List missing fields, ask if incomplete |
-| Research unavailable | Continue, note reduced confidence |
+| Research unavailable | Continue, mark the uncited claims |
 | Save fails | Display output, suggest manual save |
 
 ---
@@ -265,6 +255,6 @@ Bash: spur task batch-create --file decomposition.json   # bare JSON array (see
 When tools unavailable:
 
 1. **Continue** with available tools
-2. **Note** reduced confidence
+2. **Mark** claims left uncited
 3. **Suggest** manual fallback
 4. **Save** intermediate state

package/plugins/sp/skills/code-implementation/SKILL.md

@@ -169,7 +169,7 @@ that single full gate.
 
 **Worker iteration check order (task 1142 R3):**
 1. The **changed-path matrix** below (narrow behavior tests + affected consumers);
-2. the resolved `quality-gate` script in `light` mode (source-repo probe, else the installed twin) (biome on changed files, per-workspace typecheck, rule presets `rules:pre`/`rules:post`, related tests).
+2. `spur task gate <wbs> light` (biome on changed files, per-workspace typecheck, rule presets `rules:pre`/`rules:post`, related tests).
 
 Neither step takes the full project gate. The full gate is named only for the pipeline's `test` and `test-recheck` hops.
 

package/plugins/sp/skills/code-verification/SKILL.md

@@ -284,9 +284,8 @@ the deterministic Testing writer `spur task record` (section authorship never ha
 # write .spur/run/<wbs>-verdict.json (shape in references/verdict-schema.md), then:
 spur task record <wbs> --verdict-file .spur/run/<wbs>-verdict.json  # renders ## Testing, flips proven boxes
 # F96 residual sweep (observe-only): scan + fold AFTER record, under every --fix mode (0987).
-RESIDUAL=$(superskill script path sp residual-scan.mjs)
-node "$RESIDUAL" scan <wbs>
-node "$RESIDUAL" fold <wbs>
+spur task residual <wbs> scan
+spur task residual <wbs> fold
 # Downgrade only: re-record so Testing carries the final (folded) verdict.
 [ "$(jq -r .verdict .spur/run/<wbs>-verdict.json)" = PASS ] \
   || spur task record <wbs> --verdict-file .spur/run/<wbs>-verdict.json
@@ -297,10 +296,9 @@ node "$RESIDUAL" fold <wbs>
 > `residual-sweep` check in place (PASS → PARTIAL when a blocking residual exists), and the
 > downgrade-only re-record transcribes it — a manual verify cannot certify what the pipeline would
 > reject. The sweep runs after `record` (same order as the pipeline, 0983/0987): before it, a
-> verdict-proven box is still unticked and a clean PASS would fold to PARTIAL. Resolve the
-> script via `superskill script path sp residual-scan.mjs`; shipped surfaces never reference
-> `plugins/sp/scripts/` directly (script-contract-check rule 4). When the task reaches `done`
-> through `--next`, run `residual-scan settle` (links follow-up tasks; best-effort).
+> verdict-proven box is still unticked and a clean PASS would fold to PARTIAL. The sweep is the
+> `spur task residual` verb (no plugin script to resolve). When the task reaches `done` through
+> `--next`, run `spur task residual <wbs> settle` (files the follow-up task; best-effort).
 
 > **Corrections: the answer file is the source of truth.** `spur task record` re-transcribes
 > `## Testing` from the verdict artifact — direct `--section Testing` writes are futile. Fix
@@ -447,11 +445,13 @@ gate is **active**, evaluate feature AC satisfaction via the existing CLI.
 
    ```bash
    SPUR_BIN="${SPUR_BIN:-$([ -f apps/cli/src/index.ts ] && echo 'bun apps/cli/src/index.ts' || echo spur)}"
-   $SPUR_BIN feature check <featureId> --json
+   $SPUR_BIN feature check <featureId> --as done --strict --json
    $SPUR_BIN task list --feature <featureId> --json
    ```
 
 3. Classify **Shippable: PASS** only if **all** of:
+   - Strict completion exits 0: PASS feature-verification receipts and required dogfood evidence.
+     Projection changes no status; active-state checks are insufficient.
    - No finding whose code/message indicates **linked but unverified** scenarios
      (`L4_SCENARIO_UNVERIFIED`).
    - No **orphan / uncovered** feature scenarios (`L4_ORPHAN_SCENARIOS`,

package/plugins/sp/skills/code-verification/references/secu-review.md

@@ -83,16 +83,16 @@ verdict artifact so quality failures are not lost behind a requirements-only PAS
 Before declaring a task `done`, run this lightweight checklist. It catches the most common oversights that survive the formal pipeline gates:
 
 - [ ] Check receipt `.spur/run/<wbs>-check-receipt.json` reports `reuse: true` for the current
-      digest via `quality-gate.ts status`; run `bun run spur-check` only when it reports stale or
+      digest via `spur task gate <wbs> status`; run `bun run spur-check` only when it reports stale or
       missing (0940).
 - [ ] No `TODO` or `FIXME` without a linked task WBS.
 - [ ] `git status` shows only intentional changes (no debug artifacts, no temp files).
 - [ ] No `console.log` / `console.error` in production code (use the project logger).
 - [ ] No `--no-verify`, `--force`, or new suppression comments added to bypass gates.
 - [ ] Solution section contains `file:line` citations for every changed file.
-- [ ] API/library claims in changed docs carry a source citation (URL or `path:line`) and an honest
-      confidence level (HIGH/MEDIUM/LOW) per source-driven-development; the deterministic advisory backstop is
-      the `cited-version-strings` / `cited-api-claims` / `cited-api-claims-confidence` rules
+- [ ] API/library claims in changed docs carry a source citation (URL or `path:line`) per
+      source-driven-development; no confidence label is required (ADR-138). The deterministic
+      advisory backstop is the `cited-version-strings` / `cited-api-claims` rules
       (`.spur/rules/docs/sourced-api-claims.yaml`, run ad hoc — not wired into a preset).
 - [ ] Review section has P1–P4 findings table (even if all rows are empty — the table itself proves review happened).
 

package/plugins/sp/skills/code-verification/references/verdict-schema.md

@@ -170,6 +170,14 @@ before the proof capture (task 1109), but it qualifies **unique** basenames only
 basename (two tracked files share it) is reported and left untouched, and line drift and
 `L4.anchor-subject-mismatch` are never repaired by that pass — those remain the author's rows.
 
+A **workspace-relative** path fails the same way, and is the easier mistake to make because the
+command it came from is real. `(cd apps/cli && bun test tests/commands/builder/pr.test.ts)` is the
+right way to run the suite, but `tests/commands/builder/pr.test.ts` does not resolve from the repo
+root, so the AC-coverage guard reports `missing: <AC> — <path>` and refuses `done` (observed
+2026-10-16, task 1215 — the repair was a fresh verify pass over the whole answer). Cite both halves:
+the workspace-rooted command you ran, then the repo-root-relative anchor of the file it names
+(`apps/cli/tests/commands/builder/pr.test.ts`). One anchor per cited file per cell is enough.
+
 ## Source citations for external claims (task 1070)
 
 A Requirement/AC evidence cell that asserts an external API/library claim must carry a verifiable
@@ -185,7 +193,7 @@ citation in the SAME cell, or lint rejects the row with `evidence-citation` (blo
   presence-only — which citation proves which claim is not deterministically decidable.
 - The same two predicates (`hasClaimMarker` / `hasCitationForm` in
   `packages/app/src/services/verify-answer-lint.ts`) gate the review report's findings table through
-  `residual-scan review-gate` as the blocking `review-contract` category — one definition of
+  `spur task residual <wbs> review-gate` as the blocking `review-contract` category — one definition of
   "external claim" and one set of accepted citation forms for both paths.
 
 ## Hedged evidence is rejected on MET rows

package/plugins/sp/skills/daily-summary/SKILL.md

@@ -52,13 +52,13 @@ The daily-summary skill:
 
 ```bash
 # Generate summary for today
-node "$(superskill script path sp daily-summary/daily-summary.mjs)"
+spur history summary
 
 # Generate summary for a specific date
-node "$(superskill script path sp daily-summary/daily-summary.mjs)" --date 2026-04-17
+spur history summary --date 2026-04-17
 
 # Show summary without writing file
-node "$(superskill script path sp daily-summary/daily-summary.mjs)" --dry-run
+spur history summary --dry-run
 ```
 
 ## Workflows
@@ -165,7 +165,6 @@ Read the skill file and follow the workflow manually.
 
 ## Additional Resources
 
-- **Script source:** [scripts/daily-summary/daily-summary.ts](../../scripts/daily-summary/daily-summary.ts) — CLI implementation
-- **Tests:** [tests/daily-summary/daily-summary.test.ts](../../tests/daily-summary/daily-summary.test.ts) — unit coverage for parsing, date ranges, markdown output
+- **CLI:** `spur history summary [--date d] [--dry-run] [--output p] [--no-git] [--no-ccusage]` — see `sp:spur-cli` (history reference)
 - **Related skills:** `sp:dev-handover` (blocker handoff), `sp:dev-changelog` (commit-based changelog), `sp:spur-cli` (task management)
 - **Upstream CLI:** [ccusage](https://github.com/ryoppippi/ccusage) — AI agent token usage reporter