pantheon-opencode 2.1.4 → 2.1.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/AGENTS.global.md CHANGED
@@ -1,7 +1,7 @@
1
1
  <!-- pantheon:generated:begin — Pantheon-managed region; regenerated on install/upgrade, never edit inside -->
2
2
  <!-- Generated by scripts/build-agents-md.mjs — do not edit directly. -->
3
3
  <!-- Source of truth: src/instructions/*.instructions.md + the static sections in this script. -->
4
- <!-- pantheon:agents-md v2 artifact=global generated-at=2026-11-14T10:35:21-03:00 sha=0200775 -->
4
+ <!-- pantheon:agents-md v2 artifact=global generated-at=2026-11-14T14:44:27-03:00 sha=053433e -->
5
5
 
6
6
  # Pantheon Agent System — OpenCode
7
7
 
@@ -90,11 +90,11 @@ Required: position, reasoning, trade_offs, risks, confidence. Optional `agreemen
90
90
 
91
91
  Write one focused test per behavior; do not duplicate coverage.
92
92
 
93
- - **Test Slicing**: In dev loops, run ONLY the narrowest slice (<10s) for changed code (`node --test <slice>`, `pytest -k`); avoid unconstrained suites.
94
- - **Repo Selector & Receipt**: Prefer repo selectors (`test:quick`, `test:slice`); record command, targets, duration, and selection reason in `subtask_summary`.
95
- - **Failure/Fix Loops**: Rerun ONLY failed checks; never rerun passing tests without code changes.
96
- - **Mapping Gaps**: If a file lacks tests, run static check fallback plus nearest slice.
97
- - Slow install/E2E suites are opt-in, never per edit; reserve for release gates.
93
+ - **Risk-Based Slicing**: Run narrowest slice (<10s) first; broaden only on cross-module, contract, auth, or DB risk.
94
+ - **Suite Escalation**: Full suites are for CI/release; state if execution is agent recommendation or mandatory gate.
95
+ - **Failure/Fix Loops**: Rerun ONLY failed checks; never repeat passing tests. Never delete tests to speed up checks.
96
+ - **Mapping Gaps & Receipts**: Missing tests use static check plus nearest slice; record command, duration, and pending gates.
97
+ - **CLI vs Code**: Run CLI binaries (`opencode-quota`, `git`, `npm`) via `shell`/`bash`, not eval; on failure, never glob `~/.config`.
98
98
  - Test economy does NOT weaken stronger gates: auth, security, data-integrity, and required coverage.
99
99
 
100
100
  <!-- Source: src/instructions/yagni.instructions.md -->
package/AGENTS.md CHANGED
@@ -1,7 +1,7 @@
1
1
  <!-- pantheon:generated:begin — Pantheon-managed region; regenerated on install/upgrade, never edit inside -->
2
2
  <!-- Generated by scripts/build-agents-md.mjs — do not edit directly. -->
3
3
  <!-- Source of truth: project sections in scripts/build-agents-md.mjs. -->
4
- <!-- pantheon:agents-md v2 artifact=project generated-at=2026-11-14T10:35:21-03:00 sha=0200775 -->
4
+ <!-- pantheon:agents-md v2 artifact=project generated-at=2026-11-14T14:44:27-03:00 sha=053433e -->
5
5
 
6
6
  # Pantheon — Repository Setup
7
7
 
package/CHANGELOG.md CHANGED
@@ -13,6 +13,37 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
13
13
  `node scripts/versioning.mjs apply --notes` pre-fills the generated sections from conventional commits.
14
14
  Running apply will move this section to a versioned entry and reset the template below. -->
15
15
 
16
+ ## [v2.1.5] - 2026-11-14
17
+
18
+ ### 🆕 What's New
19
+
20
+ - **Risk-Based Validation (#444)**: Protocol distinguishes focused test slices, package suites, and mandatory repository CI gates.
21
+ - **Tool Execution Fallback**: Added `execute` alias executing commands via bash, preventing JS eval syntax errors on CLI commands.
22
+ - **CLI Execution Mandate**: Enforced CLI execution via `shell`/`bash` for binaries like `opencode-quota` with anti-panic search ban.
23
+ - **Scope Transparency (#444)**: `subtask_summary` explicitly reports validation scope and deferred pending repository CI pipeline gates.
24
+ - **Unified Skill Discovery**: Dynamic filesystem indexing of local and global skills; Zeus handoff passes explicit skill file paths.
25
+
26
+ ### Added
27
+
28
+ - **tools** — added `execute` fallback alias executing workspace commands and scripts via `/bin/bash` in `src/pantheon/tools.ts`.
29
+ - **permissions** — whitelisted `opencode-quota` in `SAFE_DEV_BINARIES` and installer bash permission rules.
30
+ - **skills** — added dynamic filesystem scanner for `.opencode/skills` and `.agents/skills` returning real path and scope.
31
+
32
+ ### Changed
33
+
34
+ - **tests (#444)** — enhanced `test-slice.mjs` with risk escalation triggers, validation scope reporting, and pending CI gate receipts.
35
+ - **instructions (#444)** — updated `test-economy` and `test-slicing` with risk-based validation protocol, anti-panic rules, and CLI mandate.
36
+ - **agents** — updated Zeus and specialist agent prompts in `roster.json` to mandate `SKILL.md` inspection before implementation.
37
+
38
+ ### ✅ Closed Issues
39
+
40
+ - #444 - define risk-based validation protocol for pantheon agents
41
+
42
+ &lt;!-- Add new changes here.
43
+ Canonical sections (docs/RELEASING.md): ### 🆕 What's New (curated highlights, ≤6 bullets ≤140 chars) →
44
+ ### Added / Changed / Deprecated / Removed / Fixed / Security → ### ⚠️ Known Issues / ### ✅ Closed Issues.
45
+ `node scripts/versioning.mjs apply --notes` pre-fills the generated sections from conventional commits.
46
+ Running apply will move this section to a versioned entry and reset the template below. --&gt;
16
47
  ## [v2.1.4] - 2026-11-14
17
48
 
18
49
  ### 🆕 What's New
package/README.md CHANGED
@@ -225,6 +225,15 @@ Working runnable reference projects demonstrating Pantheon orchestration in prac
225
225
 
226
226
  ---
227
227
 
228
+ ## What's New in v2.1.5
229
+
230
+ - **Risk-Based Validation Protocol (#444):** Explicit multi-tier protocol distinguishing focused test slices (<10s), package integration suites, and mandatory repository CI gates. Escalation to broad suites requires explicit triggers (cross-module, breaking contracts, auth, DB migrations).
231
+ - **Tool Execution Resilience & Fallback:** Added `execute` fallback tool alias executing commands directly in `/bin/bash`, preventing `SyntaxError: Assigning to rvalue` when models call `execute` with CLI commands or scripts.
232
+ - **CLI Execution Mandate:** Enforced CLI execution via `shell`/`bash` for binaries like `opencode-quota` with an anti-panic ban on speculative recursive `glob`/`grep` in `~/.config`.
233
+ - **Validation Scope Transparency (#444):** `subtask_summary` observable receipts report `validation_scope` and deferred `pending_validation` repository CI pipeline gates.
234
+
235
+ ---
236
+
228
237
  ## What's New in v2.1.4
229
238
 
230
239
  - **Global Install Shadowing & PATH Hygiene (#424):** Doctor check (H2b) detects multiple global binaries, dangling symlinks, and prefix skew. Version skew validation prevents `_npx` from binding to older global packages.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "pantheon-opencode",
3
- "version": "2.1.4",
3
+ "version": "2.1.5",
4
4
  "description": "Multi-agent orchestration for OpenCode \u2014 14 specialized agents, TDD, Themis review gates.",
5
5
  "type": "module",
6
6
  "workspaces": [
package/plugin.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "pantheon",
3
3
  "publisher": "ils15",
4
- "version": "2.1.4",
4
+ "version": "2.1.5",
5
5
  "displayName": "Pantheon",
6
6
  "description": "Conductor-Delegate orchestration platform with 14 specialized agents, 25 skills, and OpenCode support.",
7
7
  "agents": "./src/agents",
package/pyproject.toml CHANGED
@@ -4,7 +4,7 @@ build-backend = "setuptools.backends._legacy:_Backend"
4
4
 
5
5
  [project]
6
6
  name = "pantheon"
7
- version = "2.1.4"
7
+ version = "2.1.5"
8
8
  description = "Multi-agent orchestration framework"
9
9
  requires-python = ">=3.11"
10
10
  # Python sources under src/mcp/ import fastmcp/pydantic at runtime, but those
@@ -200,6 +200,15 @@ export function mapFileToSlice(filePath, cwd = process.cwd()) {
200
200
  export function resolveTestSlice(files = [], options = {}) {
201
201
  const cwd = options.cwd || process.cwd()
202
202
  const failedOnly = options.failedOnly
203
+ const trigger =
204
+ options.trigger ||
205
+ (options.crossModule
206
+ ? 'cross-module'
207
+ : options.security
208
+ ? 'security-boundary'
209
+ : options.migration
210
+ ? 'schema-migration'
211
+ : undefined)
203
212
 
204
213
  // Filter for relevant source or test files
205
214
  const relevantFiles = files.filter((f) => {
@@ -207,8 +216,19 @@ export function resolveTestSlice(files = [], options = {}) {
207
216
  return SOURCE_EXTENSIONS.has(ext) || TEST_EXTENSIONS.has(ext) || f.includes('.test.')
208
217
  })
209
218
 
210
- // 1. If repo has a predefined focused script and multiple files changed, prefer it
219
+ // 1. If an explicit escalation trigger is present and repo has a suite selector, escalate (Issue #444)
211
220
  const repoScript = findRepoSliceScript(cwd)
221
+ if (trigger && repoScript) {
222
+ return {
223
+ command: repoScript.command,
224
+ type: 'package-suite-escalation',
225
+ selection: `package suite (escalation trigger: ${trigger})`,
226
+ files: relevantFiles,
227
+ scopeJustification: `Escalated from local slice due to ${trigger}`,
228
+ }
229
+ }
230
+
231
+ // 2. If repo has a predefined focused script and multiple files changed, prefer it
212
232
  if (repoScript && (relevantFiles.length === 0 || relevantFiles.length > 2)) {
213
233
  return {
214
234
  command: repoScript.command,
@@ -299,6 +319,16 @@ export function executeTestSlice(slicePlan, options = {}) {
299
319
  const passLines = output.split('\n').filter((line) => /✔|PASS|passed/i.test(line))
300
320
  const targets = Math.max(1, passLines.length)
301
321
 
322
+ const validationScope =
323
+ options.validationScope ||
324
+ (slicePlan.type === 'package-suite-escalation'
325
+ ? 'package_suite'
326
+ : slicePlan.type === 'repo-selector'
327
+ ? 'package_suite'
328
+ : 'focused_slice')
329
+ const pendingValidation =
330
+ options.pendingValidation || 'none (CI gates deferred to repository pipeline)'
331
+
302
332
  const receipt = {
303
333
  status: isPass ? 'PASS' : 'FAIL',
304
334
  command: slicePlan.command,
@@ -306,11 +336,13 @@ export function executeTestSlice(slicePlan, options = {}) {
306
336
  durationSec,
307
337
  durationMs,
308
338
  selection: slicePlan.selection,
339
+ validationScope,
340
+ pendingValidation,
309
341
  exitCode: result.status,
310
342
  output: isPass ? output.trim().split('\n').slice(-5).join('\n') : output,
311
343
  }
312
344
 
313
- const summaryLine = `**tests:** [${receipt.status}] \`${receipt.command}\` (targets: ${receipt.targets}, duration: ${receipt.durationSec}s, selection: ${receipt.selection})`
345
+ const summaryLine = `**tests:** [${receipt.status}] \`${receipt.command}\` (targets: ${receipt.targets}, duration: ${receipt.durationSec}s, selection: ${receipt.selection})\n**validation_scope:** ${validationScope}\n**pending_validation:** ${pendingValidation}`
314
346
 
315
347
  return { ...receipt, summaryLine }
316
348
  }
package/skills-lock.json CHANGED
@@ -34,7 +34,7 @@
34
34
  "src/skills/tdd-with-agents/references/advanced-testing-patterns.md": "4aac82748bb670a61d67732168f8a265d5e66895a9183029e3da03de8e024788",
35
35
  "src/skills/tdd-with-agents/references/testing-by-layer.md": "545291301661e57d846b2db2acca1edb1219a102dce4aad7e0c3cc1a250b570a",
36
36
  "src/skills/tdd-with-agents/SKILL.md": "107495b0473a561dd570d2c25420329d68712776524a736ccc415171f7aa6c00",
37
- "src/skills/test-slicing/SKILL.md": "7361f197ab6e59a55003eb571c294f1a07388ee1d3f96b1dbfb69a84276b1bb5",
37
+ "src/skills/test-slicing/SKILL.md": "3d6f78719ee7d426f8082fd703c26a0433804c60fd3cd181d3a6886c4e94075b",
38
38
  "src/skills/verification-planning/SKILL.md": "6a3c6cb5bd5076862d369885bf0b3ba086b625495fe7a42f3f5116d07b470f3e",
39
39
  "src/skills/visual-review-pipeline/references/findings-schema.md": "776511757c0d8387fab232fced8ac9ce031a03f62b6ca539e3c38b0d45035c8e",
40
40
  "src/skills/visual-review-pipeline/references/workflow-steps.md": "13796a32076dd4d04d37a50927169929ba44a332302bd98f72ec61482ceedc09",
@@ -18,6 +18,7 @@ permission:
18
18
  "*": deny
19
19
  todowrite: allow
20
20
  todoread: allow
21
+ execute: allow
21
22
  mcp_tools:
22
23
  pantheon-resources: all
23
24
  pantheon-memory: [memory_search]
@@ -90,7 +91,8 @@ You are a frontend implementation specialist. You BUILD UI. You do NOT design ar
90
91
 
91
92
  ### Before Implementation
92
93
  1. Inspect relevant component/test/API files directly using `grep`, `glob`, and `read`; when broad or unfamiliar discovery materially reduces risk, send the discovery need to Zeus for routing
93
- 2. Read relevant frontend standards; load `skill: visual-review-pipeline` only when visual behavior changed
94
+ 2. Read relevant frontend standards; load `skill: visual-review-pipeline`
95
+ 3. When the handoff brief names an installed skill (or relevant project skill exists in `.opencode/skills/` or `.agents/skills/`), read its `SKILL.md` before writing code only when visual behavior changed
94
96
  3. Plan component/data flow when scope or risk warrants it; do not require a design phase for a bounded fix
95
97
  4. For a supplied image/screenshot, use the original reference and Zeus's visual notes as acceptance context; call `vision_describe` if the image is addressable but not directly visible. One pass per image: `vision_describe` for visual review, `vision_ocr` only when text extraction is the goal — never two vision calls on the same screenshot (each call is real gateway spend). Inspect the current state first and preserve unrelated existing work—do not delete or revert it just to match the reference unless that replacement is explicitly in scope.
96
98
  - Investigate beyond the assigned context only when necessary to resolve the task or when broader discovery materially reduces risk; verify changed behavior proportionally to its scope and risk, using focused checks for bounded changes and broader checks only when integration boundaries or risk warrant them.
@@ -17,6 +17,7 @@ permission:
17
17
  "*": deny
18
18
  todowrite: allow
19
19
  todoread: allow
20
+ execute: deny
20
21
  mcp_tools:
21
22
  pantheon-resources: all
22
23
  pantheon-memory:
@@ -19,6 +19,7 @@ permission:
19
19
  "*": deny
20
20
  apollo: allow
21
21
  todoread: allow
22
+ execute: deny
22
23
  mcp_tools:
23
24
  pantheon-resources: all
24
25
  pantheon-memory:
@@ -16,6 +16,7 @@ permission:
16
16
  "*": deny
17
17
  todowrite: allow
18
18
  todoread: allow
19
+ execute: allow
19
20
  mcp_tools:
20
21
  pantheon-resources: all
21
22
  pantheon-memory: [memory_search]
@@ -81,6 +82,7 @@ A child session has a parent and must not dispatch another Pantheon agent. Inspe
81
82
  ### Before Migration
82
83
  1. Inspect relevant schema/migrations directly; ask Zeus to route any needed broad discovery to @apollo when the schema is broad or unfamiliar enough that parallel discovery reduces risk
83
84
  2. Read the relevant model/migration context to understand relationships
85
+ 3. When a skill is specified in the brief or exists in `.opencode/skills/` (e.g. database/ORM conventions), read its `SKILL.md` before generating migrations
84
86
  3. Plan migration: what changes, impact on existing data, rollback strategy
85
87
 
86
88
  ### Migration Development (TDD)
@@ -16,6 +16,7 @@ permission:
16
16
  "*": deny
17
17
  todowrite: allow
18
18
  todoread: allow
19
+ execute: deny
19
20
  mcp_tools:
20
21
  pantheon-resources: all
21
22
  pantheon-memory: [memory_recall]
@@ -15,6 +15,7 @@ permission:
15
15
  "*": deny
16
16
  todowrite: allow
17
17
  todoread: allow
18
+ execute: allow
18
19
  mcp_tools:
19
20
  pantheon-resources: all
20
21
  pantheon-memory: [memory_search]
@@ -17,6 +17,7 @@ permission:
17
17
  apollo: allow
18
18
  todowrite: allow
19
19
  todoread: allow
20
+ execute: allow
20
21
  mcp_tools:
21
22
  pantheon-resources: all
22
23
  pantheon-memory: [memory_search]
@@ -163,7 +164,7 @@ See `skill: tdd-with-agents` for the full TDD cycle.
163
164
 
164
165
  ## Implementation & Verification Workflow
165
166
 
166
- 1. **Focused Context**: Inspect targeted files directly using `grep`, `glob`, and `read`.
167
+ 1. **Focused Context**: Inspect targeted files directly using `grep`, `glob`, and `read`. When the brief specifies a skill (or a relevant project skill exists in `.opencode/skills/`), **read its `SKILL.md` first** before implementing.
167
168
  2. **Implementation**: Implement clean async endpoints and services using Pydantic schemas, dependency injection, and parameterized queries.
168
169
  3. **Verification**: Run focused unit tests via `bash` using test slicing (e.g. `pytest tests/test_item.py -k test_create -q --tb=short`). Never run full unbounded repository suites during implementation loops.
169
170
  4. **Dependencies**: Use modern stdlib equivalents (`zoneinfo`, `tomllib`) where applicable; pin stable LTS versions.
@@ -20,6 +20,7 @@ permission:
20
20
  "*": deny
21
21
  todowrite: allow
22
22
  todoread: allow
23
+ execute: deny
23
24
  mcp_tools:
24
25
  pantheon-resources: all
25
26
  pantheon-memory: [memory_recall]
@@ -20,6 +20,7 @@ permission:
20
20
  "*": deny
21
21
  todowrite: allow
22
22
  todoread: allow
23
+ execute: deny
23
24
  mcp_tools:
24
25
  pantheon-resources: all
25
26
  pantheon-memory:
package/src/agents/nyx.md CHANGED
@@ -16,6 +16,7 @@ permission:
16
16
  "*": deny
17
17
  todowrite: allow
18
18
  todoread: allow
19
+ execute: allow
19
20
  mcp_tools:
20
21
  pantheon-resources: all
21
22
  pantheon-memory: [memory_search]
@@ -16,6 +16,7 @@ permission:
16
16
  "*": deny
17
17
  todowrite: allow
18
18
  todoread: allow
19
+ execute: allow
19
20
  mcp_tools:
20
21
  pantheon-resources: all
21
22
  pantheon-memory: [memory_search]
@@ -15,6 +15,7 @@ permission:
15
15
  "*": deny
16
16
  todowrite: allow
17
17
  todoread: allow
18
+ execute: allow
18
19
  mcp_tools:
19
20
  pantheon-resources: all
20
21
  pantheon-memory: [memory_recall]
@@ -16,6 +16,7 @@ permission:
16
16
  "*": deny
17
17
  todowrite: allow
18
18
  todoread: allow
19
+ execute: allow
19
20
  mcp_tools:
20
21
  pantheon-resources: all
21
22
  pantheon-memory: [memory_search]
@@ -19,6 +19,7 @@ permission:
19
19
  todoread: allow
20
20
  task:
21
21
  "*": allow
22
+ execute: deny
22
23
  mcp_tools:
23
24
  pantheon-resources: all
24
25
  pantheon-memory: [memory_search]
@@ -81,6 +82,7 @@ skills:
81
82
  - **DO NOT invoke @athena.** DO NOT force planning or ask for plan approval on direct coding requests.
82
83
  - **DO NOT invoke Council synthesis.** Council is reserved strictly for `/pantheon-council` or explicit multi-perspective trade-off requests.
83
84
  - When the specialist returns, report the concise deliverable and test result directly to the user.
85
+ - **Skill-Informed Handoff**: When a task involves an installed skill (built-in Pantheon skill or external in `.opencode/skills/` / `.agents/skills/`), Zeus includes `Skill: <name> (Path: .opencode/skills/<name>/SKILL.md)` in the brief and instructs the specialist to read `SKILL.md` before implementation.
84
86
  - **Asynchronous parallel subagents**: When running long test suites, heavy builds, or independent exploratory research, use `task(..., background: true)` where supported by the runtime so the main workflow does not block.
85
87
  - **Planning via @athena is strictly gated**:
86
88
  - ONLY invoke Athena when the user explicitly runs `/pantheon-deepwork` or explicitly requests a plan in the prompt ("plan this first", "create a plan", "planeje").
@@ -125,7 +127,7 @@ These rules are economic, measured against the 2026-11-05 usage audit. They appl
125
127
  * Granular choice: `"Selecionar skills específicas para instalar"` (or checkbox list if multiselect).
126
128
  * Skip choice: `"Prosseguir sem instalar skills adicionais"`.
127
129
  - Automated Specialist Execution & Batching: When the user approves in the question prompt, do NOT require them to run commands. Immediately dispatch an execution specialist with bash permitted (@hermes or @talos) to execute `npx skills add <skill1> [skill2...] -a opencode -y` in the project root (batching multiple skills in a single run).
128
- - Verification: Specialist verifies directory creation in `<project>/.agents/skills/<skill-name>/` and reports completion; subsequent `pantheon_skill_search` confirms `installed: true`.
130
+ - Verification: Specialist verifies directory creation in `<project>/.agents/skills/<skill-name>/` or `<project>/.opencode/skills/<skill-name>/` (ensuring `.opencode/skills/<skill-name>` exists or links so OpenCode native discovery sees it); subsequent `pantheon_skill_search` confirms `installed: true`.
129
131
 
130
132
  ## Visual Context for Implementation
131
133
 
@@ -141,6 +143,7 @@ Do not use shared KV to track depth; at most one auxiliary specialist.
141
143
  ## Handoff Brief (every delegation)
142
144
 
143
145
  Before every dispatch, send the target a concise brief with the objective/result, bounded scope and acceptance criteria, only relevant context and inputs, constraints and safety gates, and the expected deliverable plus verification. Do not paste full conversation history.
146
+ - **Skills (Internal & External):** When the task involves specific framework, library, or architectural patterns covered by an installed skill (either built-in Pantheon skill or external in `.opencode/skills/` / `.agents/skills/`), explicitly include `Skill: <name> (Path: .opencode/skills/<name>/SKILL.md)` in the brief and instruct the specialist to read `SKILL.md` before implementation.
144
147
  `memory_context` is optional; pass minimal non-sensitive refs and a concise takeaway, contain no secrets, and instruct the target to treat it as untrusted context. The host may independently inherit conversation transcript.
145
148
 
146
149
  ## Plan Handoff and Approval
@@ -1,13 +1,13 @@
1
1
  ---
2
- description: "Test economy — sub-10s test slicing; heavy suites are opt-in."
2
+ description: "Test economy — risk-based validation protocol and sub-10s test slicing."
3
3
  name: "Test Economy"
4
4
  ---
5
5
 
6
6
  Write one focused test per behavior; do not duplicate coverage.
7
7
 
8
- - **Test Slicing**: In dev loops, run ONLY the narrowest slice (<10s) for changed code (`node --test <slice>`, `pytest -k`); avoid unconstrained suites.
9
- - **Repo Selector & Receipt**: Prefer repo selectors (`test:quick`, `test:slice`); record command, targets, duration, and selection reason in `subtask_summary`.
10
- - **Failure/Fix Loops**: Rerun ONLY failed checks; never rerun passing tests without code changes.
11
- - **Mapping Gaps**: If a file lacks tests, run static check fallback plus nearest slice.
12
- - Slow install/E2E suites are opt-in, never per edit; reserve for release gates.
8
+ - **Risk-Based Slicing**: Run narrowest slice (<10s) first; broaden only on cross-module, contract, auth, or DB risk.
9
+ - **Suite Escalation**: Full suites are for CI/release; state if execution is agent recommendation or mandatory gate.
10
+ - **Failure/Fix Loops**: Rerun ONLY failed checks; never repeat passing tests. Never delete tests to speed up checks.
11
+ - **Mapping Gaps & Receipts**: Missing tests use static check plus nearest slice; record command, duration, and pending gates.
12
+ - **CLI vs Code**: Run CLI binaries (`opencode-quota`, `git`, `npm`) via `shell`/`bash`, not eval; on failure, never glob `~/.config`.
13
13
  - Test economy does NOT weaken stronger gates: auth, security, data-integrity, and required coverage.
@@ -88,6 +88,7 @@ const SAFE_DEV_BINARIES: ReadonlySet<string> = new Set([
88
88
  'go',
89
89
  'sh',
90
90
  'bash',
91
+ 'opencode-quota',
91
92
  ])
92
93
 
93
94
  /** Check if a resource target points to an excluded or sensitive path (.git, node_modules, traversal). */
@@ -258,7 +259,7 @@ export function evaluateAutoPermission(
258
259
 
259
260
  // 2. Read-only agents are strictly prohibited from mutating tools or shell
260
261
  if (READ_ONLY_AGENTS.has(agent)) {
261
- if (['edit', 'write', 'hashline_edit', 'bash', 'shell'].includes(action)) {
262
+ if (['edit', 'write', 'hashline_edit', 'bash', 'shell', 'execute'].includes(action)) {
262
263
  event.effect = 'deny'
263
264
  event.message = `Pantheon auto-permissions denied: read-only agent '${agent}' cannot execute mutating action '${action}'`
264
265
  return { decision: 'deny', reason: 'read-only-role-violation' }
@@ -283,8 +284,8 @@ export function evaluateAutoPermission(
283
284
  return { decision: 'allow', reason: 'safe-read-action' }
284
285
  }
285
286
 
286
- // 4. Shell commands
287
- if (action === 'bash' || action === 'shell') {
287
+ // 4. Shell and execution commands
288
+ if (action === 'bash' || action === 'shell' || action === 'execute') {
288
289
  const rawCmd =
289
290
  Array.isArray(event.resources) && typeof event.resources[0] === 'string'
290
291
  ? event.resources[0]
@@ -14,6 +14,9 @@
14
14
  * @module pantheon/skill-search-tool
15
15
  */
16
16
 
17
+ import { existsSync, readdirSync, readFileSync } from 'node:fs'
18
+ import { homedir } from 'node:os'
19
+ import { join } from 'node:path'
17
20
  // @ts-expect-error scripts/skill-search.mjs is a plain Node ESM script without bundled .d.ts
18
21
  import { searchSkillSources } from '../../scripts/skill-search.mjs'
19
22
  import { PANTHEON_SKILL_CATALOG } from './pantheon-decision-catalog.ts'
@@ -27,6 +30,101 @@ export interface SkillSearchResult {
27
30
  readonly installs?: number | null
28
31
  readonly url?: string
29
32
  readonly installCommand?: string | null
33
+ readonly path?: string | null
34
+ readonly scope?: 'internal' | 'project' | 'global'
35
+ }
36
+
37
+ export interface InstalledSkillRecord {
38
+ readonly id: string
39
+ readonly description: string
40
+ readonly path: string
41
+ readonly scope: 'internal' | 'project' | 'global'
42
+ }
43
+
44
+ export function extractSkillMetadata(
45
+ content: string,
46
+ fallbackName: string,
47
+ ): { name: string; description: string } {
48
+ let name = fallbackName
49
+ let description = ''
50
+ const fmMatch = content.match(/^---\r?\n([\s\S]*?)\r?\n---/)
51
+ const yaml = fmMatch?.[1]
52
+ if (yaml) {
53
+ const nameMatch = yaml.match(/^name:\s*["']?([^"'\r\n]+)["']?/m)
54
+ if (nameMatch?.[1]) name = nameMatch[1].trim()
55
+ const descMatch = yaml.match(/^description:\s*["']?([^"'\r\n]+)["']?/m)
56
+ if (descMatch?.[1]) description = descMatch[1].trim()
57
+ }
58
+ if (!description) {
59
+ const withoutFm = content.replace(/^---\r?\n[\s\S]*?\r?\n---/, '').trim()
60
+ const lines = withoutFm.split('\n')
61
+ for (const l of lines) {
62
+ const trimmed = l.trim()
63
+ if (trimmed && !trimmed.startsWith('#')) {
64
+ description = trimmed.slice(0, 160)
65
+ break
66
+ }
67
+ }
68
+ }
69
+ return { name, description }
70
+ }
71
+
72
+ export function scanInstalledSkills(
73
+ cwd: string = process.cwd(),
74
+ ): Map<string, InstalledSkillRecord> {
75
+ const map = new Map<string, InstalledSkillRecord>()
76
+
77
+ // 1. Built-in Pantheon catalog (internal base)
78
+ for (const entry of PANTHEON_SKILL_CATALOG) {
79
+ map.set(entry.id.toLowerCase(), {
80
+ id: entry.id,
81
+ description: entry.description,
82
+ path: `src/skills/${entry.id}/SKILL.md`,
83
+ scope: 'internal',
84
+ })
85
+ }
86
+
87
+ const scanDir = (baseDir: string, scope: 'project' | 'global') => {
88
+ try {
89
+ if (!existsSync(baseDir)) return
90
+ const entries = readdirSync(baseDir, { withFileTypes: true })
91
+ for (const entry of entries) {
92
+ if (!entry.isDirectory() && !entry.isSymbolicLink()) continue
93
+ const skillName = entry.name
94
+ if (skillName.startsWith('.')) continue
95
+ const skillMd = join(baseDir, skillName, 'SKILL.md')
96
+ if (existsSync(skillMd)) {
97
+ try {
98
+ const raw = readFileSync(skillMd, 'utf8')
99
+ const { name, description } = extractSkillMetadata(raw, skillName)
100
+ const id = (name || skillName).trim()
101
+ map.set(id.toLowerCase(), {
102
+ id,
103
+ description: description || skillName,
104
+ path: skillMd,
105
+ scope,
106
+ })
107
+ } catch {
108
+ // ignore unreadable files
109
+ }
110
+ }
111
+ }
112
+ } catch {
113
+ // ignore directory errors
114
+ }
115
+ }
116
+
117
+ // 2. Global directories
118
+ const xdgConfig = process.env.XDG_CONFIG_HOME || join(homedir(), '.config')
119
+ scanDir(join(xdgConfig, 'opencode', 'skills'), 'global')
120
+ scanDir(join(homedir(), '.agents', 'skills'), 'global')
121
+
122
+ // 3. Project directories (override global/internal)
123
+ scanDir(join(cwd, '.opencode', 'skills'), 'project')
124
+ scanDir(join(cwd, '.agents', 'skills'), 'project')
125
+ scanDir(join(cwd, 'skills'), 'project')
126
+
127
+ return map
30
128
  }
31
129
 
32
130
  export interface SkillSearchOutput {
@@ -74,7 +172,10 @@ const SKILL_SEARCH_INPUT_SCHEMA: Record<string, unknown> = {
74
172
  },
75
173
  }
76
174
 
77
- export function executeSkillSearch(input: Record<string, unknown>): SkillSearchOutput {
175
+ export function executeSkillSearch(
176
+ input: Record<string, unknown>,
177
+ options: { cwd?: string } = {},
178
+ ): SkillSearchOutput {
78
179
  const query = typeof input.query === 'string' ? input.query.trim() : ''
79
180
  if (query.length === 0) {
80
181
  return { ok: false, query: '', count: 0, results: [], failures: ['query must be non-empty'] }
@@ -97,7 +198,10 @@ export function executeSkillSearch(input: Record<string, unknown>): SkillSearchO
97
198
  const results: SkillSearchResult[] = []
98
199
  const failures: string[] = []
99
200
 
100
- // 1. Tier 1: Installed Pantheon catalog (in-memory, instant, zero-network)
201
+ const cwd = options.cwd ?? process.cwd()
202
+ const installedMap = scanInstalledSkills(cwd)
203
+
204
+ // 1. Tier 1: Installed skills (Pantheon catalog + local filesystem .opencode/skills and .agents/skills)
101
205
  if (sources.has('installed')) {
102
206
  const tokens = query
103
207
  .toLowerCase()
@@ -105,7 +209,7 @@ export function executeSkillSearch(input: Record<string, unknown>): SkillSearchO
105
209
  .filter((t) => t.length > 2)
106
210
 
107
211
  if (tokens.length > 0) {
108
- for (const entry of PANTHEON_SKILL_CATALOG) {
212
+ for (const entry of installedMap.values()) {
109
213
  const idLower = entry.id.toLowerCase()
110
214
  const descLower = entry.description.toLowerCase()
111
215
  const matchScore = tokens.reduce((score, t) => {
@@ -129,6 +233,8 @@ export function executeSkillSearch(input: Record<string, unknown>): SkillSearchO
129
233
  installed: true,
130
234
  installs: null,
131
235
  installCommand: null,
236
+ path: entry.path,
237
+ scope: entry.scope,
132
238
  })
133
239
  }
134
240
  }
@@ -152,15 +258,26 @@ export function executeSkillSearch(input: Record<string, unknown>): SkillSearchO
152
258
 
153
259
  if (Array.isArray(extRes.entries)) {
154
260
  for (const entry of extRes.entries) {
155
- results.push({
261
+ const bareName = (
262
+ entry.skill.includes('@') ? entry.skill.split('@')[1] : entry.skill
263
+ ).toLowerCase()
264
+ const isInstalled =
265
+ installedMap.has(entry.skill.toLowerCase()) || installedMap.has(bareName)
266
+ const installedRecord =
267
+ installedMap.get(entry.skill.toLowerCase()) || installedMap.get(bareName)
268
+
269
+ const pushItem: SkillSearchResult = {
156
270
  skill: entry.skill,
157
271
  description: entry.skill,
158
272
  source: entry.source as 'anthropic' | 'skills-sh' | 'github',
159
- installed: false,
273
+ installed: isInstalled,
160
274
  installs: entry.installs ?? null,
161
- url: entry.url,
162
- installCommand: `npx skills add ${entry.skill} -a opencode -y`,
163
- })
275
+ ...(entry.url ? { url: entry.url } : {}),
276
+ installCommand: isInstalled ? null : `npx skills add ${entry.skill} -a opencode -y`,
277
+ path: isInstalled && installedRecord ? installedRecord.path : null,
278
+ ...(isInstalled && installedRecord ? { scope: installedRecord.scope } : {}),
279
+ }
280
+ results.push(pushItem)
164
281
  }
165
282
  }
166
283
  } catch (err) {
@@ -186,8 +303,11 @@ export function createV2SkillSearchDefinition(): V2ToolDef {
186
303
  description: TOOL_DESCRIPTION,
187
304
  input: SKILL_SEARCH_INPUT_SCHEMA,
188
305
  output: {},
189
- execute: async (input: Record<string, unknown>) => ({
190
- output: JSON.stringify(executeSkillSearch(input), null, 2),
191
- }),
306
+ execute: async (input: Record<string, unknown>, context?: unknown) => {
307
+ const cwd = (context as { directory?: string } | undefined)?.directory || process.cwd()
308
+ return {
309
+ output: JSON.stringify(executeSkillSearch(input, { cwd }), null, 2),
310
+ }
311
+ },
192
312
  }
193
313
  }
@@ -567,7 +567,8 @@ const shellTool: V2ToolDef = {
567
567
  input: SHELL_INPUT,
568
568
  output: {},
569
569
  execute: async (input, context) => {
570
- const command = String((input as Record<string, unknown>).command ?? '')
570
+ const raw = (input as Record<string, unknown>) ?? {}
571
+ const command = String(raw.command ?? raw.cmd ?? '')
571
572
  if (!command.trim()) {
572
573
  return { output: '', ok: false, error: 'Command cannot be empty' }
573
574
  }
@@ -51,7 +51,8 @@ var SAFE_DEV_BINARIES = /* @__PURE__ */ new Set([
51
51
  "cargo",
52
52
  "go",
53
53
  "sh",
54
- "bash"
54
+ "bash",
55
+ "opencode-quota"
55
56
  ]);
56
57
  function isExcludedReadPath(rawPath) {
57
58
  const trimmed = rawPath.trim();
@@ -144,7 +145,7 @@ function evaluateAutoPermission(event, options = {}) {
144
145
  return { decision: "allow", reason: "pantheon-native-tool" };
145
146
  }
146
147
  if (READ_ONLY_AGENTS.has(agent)) {
147
- if (["edit", "write", "hashline_edit", "bash", "shell"].includes(action)) {
148
+ if (["edit", "write", "hashline_edit", "bash", "shell", "execute"].includes(action)) {
148
149
  event.effect = "deny";
149
150
  event.message = `Pantheon auto-permissions denied: read-only agent '${agent}' cannot execute mutating action '${action}'`;
150
151
  return { decision: "deny", reason: "read-only-role-violation" };
@@ -163,7 +164,7 @@ function evaluateAutoPermission(event, options = {}) {
163
164
  event.effect = "allow";
164
165
  return { decision: "allow", reason: "safe-read-action" };
165
166
  }
166
- if (action === "bash" || action === "shell") {
167
+ if (action === "bash" || action === "shell" || action === "execute") {
167
168
  const rawCmd = Array.isArray(event.resources) && typeof event.resources[0] === "string" ? event.resources[0] : "";
168
169
  if (rawCmd) {
169
170
  if (isDangerousCommand(rawCmd)) {
@@ -13799,6 +13800,11 @@ function createModelCommand(options = {}) {
13799
13800
  };
13800
13801
  }
13801
13802
 
13803
+ // src/pantheon/skill-search-tool.ts
13804
+ import { existsSync as existsSync8, readdirSync as readdirSync2, readFileSync as readFileSync7 } from "node:fs";
13805
+ import { homedir as homedir8 } from "node:os";
13806
+ import { join as join18 } from "node:path";
13807
+
13802
13808
  // scripts/skill-search.mjs
13803
13809
  import { spawnSync } from "node:child_process";
13804
13810
  import { pathToFileURL } from "node:url";
@@ -14018,6 +14024,75 @@ var invokedDirectly = process.argv[1] !== void 0 && import.meta.url === pathToFi
14018
14024
  if (invokedDirectly) main();
14019
14025
 
14020
14026
  // src/pantheon/skill-search-tool.ts
14027
+ function extractSkillMetadata(content, fallbackName) {
14028
+ let name = fallbackName;
14029
+ let description = "";
14030
+ const fmMatch = content.match(/^---\r?\n([\s\S]*?)\r?\n---/);
14031
+ const yaml = fmMatch?.[1];
14032
+ if (yaml) {
14033
+ const nameMatch = yaml.match(/^name:\s*["']?([^"'\r\n]+)["']?/m);
14034
+ if (nameMatch?.[1]) name = nameMatch[1].trim();
14035
+ const descMatch = yaml.match(/^description:\s*["']?([^"'\r\n]+)["']?/m);
14036
+ if (descMatch?.[1]) description = descMatch[1].trim();
14037
+ }
14038
+ if (!description) {
14039
+ const withoutFm = content.replace(/^---\r?\n[\s\S]*?\r?\n---/, "").trim();
14040
+ const lines = withoutFm.split("\n");
14041
+ for (const l of lines) {
14042
+ const trimmed = l.trim();
14043
+ if (trimmed && !trimmed.startsWith("#")) {
14044
+ description = trimmed.slice(0, 160);
14045
+ break;
14046
+ }
14047
+ }
14048
+ }
14049
+ return { name, description };
14050
+ }
14051
+ function scanInstalledSkills(cwd = process.cwd()) {
14052
+ const map = /* @__PURE__ */ new Map();
14053
+ for (const entry of PANTHEON_SKILL_CATALOG) {
14054
+ map.set(entry.id.toLowerCase(), {
14055
+ id: entry.id,
14056
+ description: entry.description,
14057
+ path: `src/skills/${entry.id}/SKILL.md`,
14058
+ scope: "internal"
14059
+ });
14060
+ }
14061
+ const scanDir = (baseDir, scope) => {
14062
+ try {
14063
+ if (!existsSync8(baseDir)) return;
14064
+ const entries = readdirSync2(baseDir, { withFileTypes: true });
14065
+ for (const entry of entries) {
14066
+ if (!entry.isDirectory() && !entry.isSymbolicLink()) continue;
14067
+ const skillName = entry.name;
14068
+ if (skillName.startsWith(".")) continue;
14069
+ const skillMd = join18(baseDir, skillName, "SKILL.md");
14070
+ if (existsSync8(skillMd)) {
14071
+ try {
14072
+ const raw = readFileSync7(skillMd, "utf8");
14073
+ const { name, description } = extractSkillMetadata(raw, skillName);
14074
+ const id = (name || skillName).trim();
14075
+ map.set(id.toLowerCase(), {
14076
+ id,
14077
+ description: description || skillName,
14078
+ path: skillMd,
14079
+ scope
14080
+ });
14081
+ } catch {
14082
+ }
14083
+ }
14084
+ }
14085
+ } catch {
14086
+ }
14087
+ };
14088
+ const xdgConfig = process.env.XDG_CONFIG_HOME || join18(homedir8(), ".config");
14089
+ scanDir(join18(xdgConfig, "opencode", "skills"), "global");
14090
+ scanDir(join18(homedir8(), ".agents", "skills"), "global");
14091
+ scanDir(join18(cwd, ".opencode", "skills"), "project");
14092
+ scanDir(join18(cwd, ".agents", "skills"), "project");
14093
+ scanDir(join18(cwd, "skills"), "project");
14094
+ return map;
14095
+ }
14021
14096
  var TOOL_DESCRIPTION3 = "Advisory discovery tool for skills. Queries installed Pantheon skills and external registries (Anthropic catalog, skills.sh, GitHub). Returns matched skill IDs, descriptions, install counts, and provenance tiers. Zeus uses this to recommend project-scoped skills (.agents/skills/) before proactively prompting the user via the question tool.";
14022
14097
  var SKILL_SEARCH_INPUT_SCHEMA = {
14023
14098
  type: "object",
@@ -14047,7 +14122,7 @@ var SKILL_SEARCH_INPUT_SCHEMA = {
14047
14122
  }
14048
14123
  }
14049
14124
  };
14050
- function executeSkillSearch(input) {
14125
+ function executeSkillSearch(input, options = {}) {
14051
14126
  const query = typeof input.query === "string" ? input.query.trim() : "";
14052
14127
  if (query.length === 0) {
14053
14128
  return { ok: false, query: "", count: 0, results: [], failures: ["query must be non-empty"] };
@@ -14063,10 +14138,12 @@ function executeSkillSearch(input) {
14063
14138
  );
14064
14139
  const results = [];
14065
14140
  const failures = [];
14141
+ const cwd = options.cwd ?? process.cwd();
14142
+ const installedMap = scanInstalledSkills(cwd);
14066
14143
  if (sources.has("installed")) {
14067
14144
  const tokens = query.toLowerCase().split(/[^a-z0-9]+/).filter((t) => t.length > 2);
14068
14145
  if (tokens.length > 0) {
14069
- for (const entry of PANTHEON_SKILL_CATALOG) {
14146
+ for (const entry of installedMap.values()) {
14070
14147
  const idLower = entry.id.toLowerCase();
14071
14148
  const descLower = entry.description.toLowerCase();
14072
14149
  const matchScore = tokens.reduce((score, t) => {
@@ -14080,7 +14157,9 @@ function executeSkillSearch(input) {
14080
14157
  source: "installed",
14081
14158
  installed: true,
14082
14159
  installs: null,
14083
- installCommand: null
14160
+ installCommand: null,
14161
+ path: entry.path,
14162
+ scope: entry.scope
14084
14163
  });
14085
14164
  }
14086
14165
  }
@@ -14099,15 +14178,21 @@ function executeSkillSearch(input) {
14099
14178
  }
14100
14179
  if (Array.isArray(extRes.entries)) {
14101
14180
  for (const entry of extRes.entries) {
14102
- results.push({
14181
+ const bareName = (entry.skill.includes("@") ? entry.skill.split("@")[1] : entry.skill).toLowerCase();
14182
+ const isInstalled = installedMap.has(entry.skill.toLowerCase()) || installedMap.has(bareName);
14183
+ const installedRecord = installedMap.get(entry.skill.toLowerCase()) || installedMap.get(bareName);
14184
+ const pushItem = {
14103
14185
  skill: entry.skill,
14104
14186
  description: entry.skill,
14105
14187
  source: entry.source,
14106
- installed: false,
14188
+ installed: isInstalled,
14107
14189
  installs: entry.installs ?? null,
14108
- url: entry.url,
14109
- installCommand: `npx skills add ${entry.skill} -a opencode -y`
14110
- });
14190
+ ...entry.url ? { url: entry.url } : {},
14191
+ installCommand: isInstalled ? null : `npx skills add ${entry.skill} -a opencode -y`,
14192
+ path: isInstalled && installedRecord ? installedRecord.path : null,
14193
+ ...isInstalled && installedRecord ? { scope: installedRecord.scope } : {}
14194
+ };
14195
+ results.push(pushItem);
14111
14196
  }
14112
14197
  }
14113
14198
  } catch (err) {
@@ -14131,9 +14216,12 @@ function createV2SkillSearchDefinition() {
14131
14216
  description: TOOL_DESCRIPTION3,
14132
14217
  input: SKILL_SEARCH_INPUT_SCHEMA,
14133
14218
  output: {},
14134
- execute: async (input) => ({
14135
- output: JSON.stringify(executeSkillSearch(input), null, 2)
14136
- })
14219
+ execute: async (input, context) => {
14220
+ const cwd = context?.directory || process.cwd();
14221
+ return {
14222
+ output: JSON.stringify(executeSkillSearch(input, { cwd }), null, 2)
14223
+ };
14224
+ }
14137
14225
  };
14138
14226
  }
14139
14227
 
@@ -14452,7 +14540,8 @@ var shellTool = {
14452
14540
  input: SHELL_INPUT,
14453
14541
  output: {},
14454
14542
  execute: async (input, context) => {
14455
- const command = String(input.command ?? "");
14543
+ const raw = input ?? {};
14544
+ const command = String(raw.command ?? raw.cmd ?? "");
14456
14545
  if (!command.trim()) {
14457
14546
  return { output: "", ok: false, error: "Command cannot be empty" };
14458
14547
  }
@@ -2062,7 +2062,7 @@ async function readChangelogForBundle(bundleUrl) {
2062
2062
  }
2063
2063
  /** The ambient build-time version, or null outside a tsdown build. */
2064
2064
  function embeddedVersion() {
2065
- return "2.1.4";
2065
+ return "2.1.5";
2066
2066
  }
2067
2067
  /**
2068
2068
  * Resolve the running Pantheon version for the what's-new window: the plugin
@@ -2287,7 +2287,7 @@ function persistUpdateCheckState(previous, latest) {
2287
2287
  } catch {}
2288
2288
  }
2289
2289
  function resolveRunningVersion() {
2290
- return "2.1.4";
2290
+ return "2.1.5";
2291
2291
  }
2292
2292
  const VALUE_WIDTH = 24;
2293
2293
  function SettingsWindow(props) {
@@ -6513,7 +6513,7 @@ async function checkForPantheonUpdate(context) {
6513
6513
  const latest = String((await response.json())?.version ?? "");
6514
6514
  if (!latest) return;
6515
6515
  const notice = pickUpdateNotice({
6516
- currentVersion: "2.1.4",
6516
+ currentVersion: "2.1.5",
6517
6517
  latestVersion: latest,
6518
6518
  state: readState(),
6519
6519
  nowMs: Date.now()
@@ -7325,7 +7325,7 @@ function Sidebar(props) {
7325
7325
  get children() {
7326
7326
  var _el$13 = createElement("text");
7327
7327
  setProp(_el$13, "attributes", 1);
7328
- insert(_el$13, `Pantheon v2.1.4`);
7328
+ insert(_el$13, `Pantheon v2.1.5`);
7329
7329
  effect((_$p) => setProp(_el$13, "fg", text(), _$p));
7330
7330
  return _el$13;
7331
7331
  }
@@ -8097,7 +8097,7 @@ function HomeStatus(props) {
8097
8097
  },
8098
8098
  get children() {
8099
8099
  var _el$73 = createElement("text");
8100
- insert(_el$73, () => `Pantheon v2.1.4${preset().name ? ` ▸ ${preset().name}` : ""}`);
8100
+ insert(_el$73, () => `Pantheon v2.1.5${preset().name ? ` ▸ ${preset().name}` : ""}`);
8101
8101
  effect((_$p) => setProp(_el$73, "fg", muted(), _$p));
8102
8102
  return _el$73;
8103
8103
  }
@@ -1,12 +1,12 @@
1
1
  {
2
2
  "name": "pantheon-tui",
3
- "version": "2.1.4",
3
+ "version": "2.1.5",
4
4
  "lockfileVersion": 3,
5
5
  "requires": true,
6
6
  "packages": {
7
7
  "": {
8
8
  "name": "pantheon-tui",
9
- "version": "2.1.4",
9
+ "version": "2.1.5",
10
10
  "license": "MIT",
11
11
  "dependencies": {
12
12
  "@opencode/plugin": "2.0.26",
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "pantheon-tui",
3
3
  "type": "module",
4
- "version": "2.1.4",
4
+ "version": "2.1.5",
5
5
  "description": "Pantheon TUI plugin \u2014 V1 rich sidebar and usage gauges; native V2 status, session and delegation sidebar",
6
6
  "license": "MIT",
7
7
  "engines": {
@@ -19,13 +19,18 @@ Always prefer the repository's own focused selector or diff-aware command over g
19
19
  2. If a project-defined focused selector exists, execute it directly.
20
20
  3. If absent, map the modified file directly to its corresponding test file.
21
21
 
22
- ### 2. Scope Escalation & Mapping Gap Protocol
23
- - **No Silent Escalation:** NEVER silently broaden a test run from a modified file to an entire package or global suite.
24
- - **Explicit Justification:** If an expansion beyond the narrow slice is genuinely warranted (e.g. cross-cutting auth change, schema migration affecting multiple handlers), state visibly before running:
25
- `[Test Scope Expansion] Scope: <files> | Selected: <tests> | Reason: <concrete risk/integration rationale>`
22
+ ### 2. Scope Escalation & Risk-Based Protocol (Issue #444)
23
+ - **Focused Slice First:** Identify affected files/behaviors and run the narrowest slice (<10s).
24
+ - **Triggers for Package-Level Expansion:** Broaden beyond the local slice ONLY when changes:
25
+ 1. Cross module/package boundaries
26
+ 2. Alter public API/contracts or data schemas/migrations
27
+ 3. Touch authentication, authorization, or security boundaries
28
+ - **Broad Suite Distinction:** When running a broad suite, state explicitly whether it is an agent recommendation or a mandatory repository/CI gate.
29
+ - **Never Delete Tests to Speed Up:** Propose test removal only with concrete evidence of duplication, flakiness, or zero detection value.
30
+ - **CLI vs Code Execution:** Execute CLI binaries (`opencode-quota`, `git`, `npm`, `pytest`) via `shell`/`bash`, never JS eval (`execute`); on failure, never glob `~/.config`.
26
31
  - **Mapping Gap Handling:** If a modified file has no direct test counterpart:
27
32
  1. Report the mapping gap explicitly: `[Test Mapping Gap] No direct test file for <file>.`
28
- 2. Execute a bounded risk-appropriate fallback (static lint/typecheck gate e.g. `npx biome check <file>` or `ruff check <file>` plus the closest integration slice).
33
+ 2. Execute a bounded risk-appropriate fallback (static lint/typecheck gate plus closest slice).
29
34
  3. DO NOT escalate to running the entire repository test suite.
30
35
 
31
36
  ### 3. Failure/Fix Loop (Rerun Only Failed Tests)
@@ -36,11 +41,12 @@ Always prefer the repository's own focused selector or diff-aware command over g
36
41
  3. **DO NOT rerun checks that already passed** when their covered code did not change.
37
42
  4. Once the targeted test passes, rerun the single slice file once to confirm no regressions within the file.
38
43
 
39
- ### 4. Observable Task Receipt Summary
40
- Every implementation specialist MUST include the detailed test receipt in the `subtask_summary`:
44
+ ### 4. Observable Task Receipt Summary (Issue #444)
45
+ Every implementation specialist MUST include the detailed test receipt in `subtask_summary`, distinguishing agent-executed tests from CI gates and pending validations:
41
46
  ```markdown
42
47
  **tests:** [PASS | WARN | FAIL] `<command>` (targets: N, duration: <X>s, selection: <reason or repo selector>)
43
- **test_scope_justification:** <required ONLY if broader than a focused slice>
48
+ **validation_scope:** focused_slice | package_suite | ci_gate_required
49
+ **pending_validation:** none | [deferred broad suites / CI gates]
44
50
  ```
45
51
  *Example:*
46
52
  `**tests:** [PASS] npx tsx tests/pantheon/auto-permissions.test.ts (targets: 10, duration: 0.12s, selection: direct file slice for auto-permissions.ts)`