@ryuenn3123/agentic-senior-core 4.3.15 → 5.1.0
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/plugins/marketplace.json +21 -0
- package/.agents/rules/agentic-senior-core.md +126 -0
- package/.claude-plugin/marketplace.json +17 -0
- package/.claude-plugin/plugin.json +17 -0
- package/.clinerules/agentic-senior-core.md +94 -0
- package/.codex-plugin/plugin.json +24 -0
- package/.cursor/rules/agentic-senior-core.mdc +99 -0
- package/.devin-plugin/plugin.json +13 -0
- package/.github/copilot-instructions.md +94 -0
- package/.github/plugin/marketplace.json +20 -0
- package/.github/plugin/plugin.json +16 -0
- package/.kiro/steering/agentic-senior-core.md +94 -0
- package/.openclaw/skills/asc/SKILL.md +23 -0
- package/.openclaw/skills/asc-audit/SKILL.md +28 -0
- package/.openclaw/skills/asc-review/SKILL.md +54 -0
- package/.opencode/plugins/agentic-senior-core.mjs +31 -0
- package/.windsurf/rules/agentic-senior-core.md +94 -0
- package/AGENTS.md +126 -179
- package/CONTRIBUTING.md +93 -63
- package/README.md +160 -111
- package/__init__.py +61 -0
- package/bin/agentic-senior-core.js +41 -43
- package/commands/asc-audit.md +1 -0
- package/commands/asc-audit.toml +2 -0
- package/commands/asc-help.md +9 -0
- package/commands/asc-help.toml +2 -0
- package/commands/asc-refactor.md +1 -0
- package/commands/asc-refactor.toml +2 -0
- package/commands/asc-review.md +1 -0
- package/commands/asc-review.toml +2 -0
- package/gemini-extension.json +10 -0
- package/hooks/copilot-hooks.json +21 -0
- package/hooks/hooks.json +31 -0
- package/hooks/session-start.js +36 -0
- package/hooks/subagent-start.js +38 -0
- package/lib/cli/commands/adapter.mjs +109 -0
- package/lib/cli/commands/clean.mjs +63 -0
- package/lib/cli/commands/status.mjs +94 -0
- package/package.json +43 -67
- package/plugin.yaml +16 -0
- package/scripts/uninstall.js +38 -0
- package/skills/asc/SKILL.md +16 -0
- package/skills/asc-audit/SKILL.md +28 -0
- package/skills/asc-refactor/SKILL.md +35 -0
- package/skills/asc-review/SKILL.md +54 -0
- package/.agent-context/policies/llm-judge-threshold.json +0 -29
- package/.agent-context/prompts/bootstrap-design.md +0 -63
- package/.agent-context/prompts/compact-natural-mode.md +0 -100
- package/.agent-context/prompts/init-project.md +0 -60
- package/.agent-context/prompts/refactor.md +0 -38
- package/.agent-context/prompts/review-code.md +0 -31
- package/.agent-context/review-checklists/architecture-review.md +0 -57
- package/.agent-context/review-checklists/pr-checklist.md +0 -136
- package/.agent-context/rules/api-docs.md +0 -24
- package/.agent-context/rules/api-versioning.md +0 -16
- package/.agent-context/rules/architecture.md +0 -33
- package/.agent-context/rules/background-jobs.md +0 -17
- package/.agent-context/rules/config-and-flags.md +0 -16
- package/.agent-context/rules/database-design.md +0 -18
- package/.agent-context/rules/docker-runtime.md +0 -17
- package/.agent-context/rules/error-handling.md +0 -16
- package/.agent-context/rules/event-driven.md +0 -16
- package/.agent-context/rules/frontend-architecture.md +0 -54
- package/.agent-context/rules/microservices.md +0 -16
- package/.agent-context/rules/migrations.md +0 -16
- package/.agent-context/rules/observability.md +0 -16
- package/.agent-context/rules/performance.md +0 -16
- package/.agent-context/rules/realtime.md +0 -16
- package/.agent-context/rules/resilience.md +0 -18
- package/.agent-context/rules/security.md +0 -18
- package/.agent-context/rules/testing.md +0 -16
- package/.agent-context/state/README.md +0 -25
- package/.agent-context/state/architecture-map.md +0 -40
- package/.agent-context/state/benchmark-comparison-schema.json +0 -181
- package/.agent-context/state/benchmark-reproducibility.json +0 -87
- package/.agent-context/state/benchmark-thresholds.json +0 -10
- package/.agent-context/state/benchmark-writer-judge-config.json +0 -58
- package/.agent-context/state/dependency-map.md +0 -41
- package/.agent-context/state/memory-adapter-contract.json +0 -52
- package/.agent-context/state/memory-schema-v1.json +0 -88
- package/GEMINI.md +0 -1
- package/benchmarks/README.md +0 -60
- package/benchmarks/compact-natural-mode/fixtures.mjs +0 -359
- package/benchmarks/compact-natural-mode/scorer.mjs +0 -331
- package/benchmarks/runtime-token-saver/fixtures.mjs +0 -836
- package/lib/cli/adaptive-context/catalog.mjs +0 -390
- package/lib/cli/adaptive-context/file-signals.mjs +0 -100
- package/lib/cli/adaptive-context/implications.mjs +0 -44
- package/lib/cli/adaptive-context.mjs +0 -365
- package/lib/cli/audits/typography-palette-anti-repeat/color-utils.mjs +0 -156
- package/lib/cli/audits/typography-palette-anti-repeat/file-scanner.mjs +0 -103
- package/lib/cli/audits/typography-palette-anti-repeat/typography-utils.mjs +0 -70
- package/lib/cli/audits/typography-palette-anti-repeat-audit.mjs +0 -239
- package/lib/cli/backup.mjs +0 -210
- package/lib/cli/commands/audit-design-anti-repeat.mjs +0 -39
- package/lib/cli/commands/context.mjs +0 -140
- package/lib/cli/commands/init/project-context.mjs +0 -60
- package/lib/cli/commands/init/runtime-environment.mjs +0 -59
- package/lib/cli/commands/init/setup-decisions.mjs +0 -83
- package/lib/cli/commands/init.mjs +0 -654
- package/lib/cli/commands/launch.mjs +0 -76
- package/lib/cli/commands/optimize.mjs +0 -284
- package/lib/cli/commands/rollback.mjs +0 -64
- package/lib/cli/commands/upgrade/token-optimization-state.mjs +0 -51
- package/lib/cli/commands/upgrade.mjs +0 -451
- package/lib/cli/compiler.mjs +0 -570
- package/lib/cli/constants.mjs +0 -136
- package/lib/cli/detector/constants.mjs +0 -135
- package/lib/cli/detector/design-evidence/collector.mjs +0 -256
- package/lib/cli/detector/design-evidence/constants.mjs +0 -39
- package/lib/cli/detector/design-evidence/file-traversal.mjs +0 -83
- package/lib/cli/detector/design-evidence/structured-attribute-evidence.mjs +0 -117
- package/lib/cli/detector/design-evidence/summary.mjs +0 -109
- package/lib/cli/detector/design-evidence/utility-helpers.mjs +0 -122
- package/lib/cli/detector/design-evidence.mjs +0 -25
- package/lib/cli/detector/stack-detection.mjs +0 -243
- package/lib/cli/detector/ui-signals.mjs +0 -150
- package/lib/cli/detector/workspace-scan.mjs +0 -177
- package/lib/cli/detector.mjs +0 -23
- package/lib/cli/init-detection-flow.mjs +0 -48
- package/lib/cli/init-options.mjs +0 -237
- package/lib/cli/init-selection.mjs +0 -29
- package/lib/cli/memory-continuity.mjs +0 -537
- package/lib/cli/preflight.mjs +0 -76
- package/lib/cli/project-scaffolder/constants.mjs +0 -69
- package/lib/cli/project-scaffolder/discovery.mjs +0 -272
- package/lib/cli/project-scaffolder/prompt-builders.mjs +0 -171
- package/lib/cli/project-scaffolder/storage.mjs +0 -144
- package/lib/cli/project-scaffolder/ui-scope-detection.mjs +0 -36
- package/lib/cli/project-scaffolder.mjs +0 -29
- package/lib/cli/rollback.mjs +0 -66
- package/lib/cli/token-optimization.mjs +0 -486
- package/lib/cli/utils/filesystem.mjs +0 -81
- package/lib/cli/utils/managed-surface.mjs +0 -280
- package/lib/cli/utils/prompting.mjs +0 -44
- package/lib/cli/utils.mjs +0 -479
- package/mcp.json +0 -134
- package/scripts/adaptive-context/fixtures.mjs +0 -183
- package/scripts/adaptive-context-benchmark.mjs +0 -9
- package/scripts/ascx-runtime-token-saver-benchmark.mjs +0 -9
- package/scripts/audit-cache-layer-contract.mjs +0 -263
- package/scripts/audit-caching-scope-hygiene.mjs +0 -268
- package/scripts/audit-file-size.mjs +0 -219
- package/scripts/audit-reflection-citations.mjs +0 -163
- package/scripts/audit-release-bundle.mjs +0 -170
- package/scripts/audit-rule-id-uniqueness.mjs +0 -313
- package/scripts/audit-typography-palette-anti-repeat.mjs +0 -120
- package/scripts/benchmark-evidence-bundle.mjs +0 -645
- package/scripts/benchmark-gate.mjs +0 -121
- package/scripts/benchmark-intelligence.mjs +0 -33
- package/scripts/benchmark-writer-judge-matrix.mjs +0 -379
- package/scripts/build-release-benchmark-bundle.mjs +0 -202
- package/scripts/bump-version.mjs +0 -103
- package/scripts/clean-local-artifacts.mjs +0 -78
- package/scripts/compact-natural-mode-benchmark.mjs +0 -9
- package/scripts/context-triggered-audit.mjs +0 -391
- package/scripts/detection-benchmark.mjs +0 -127
- package/scripts/docs-quality-drift-report.mjs +0 -358
- package/scripts/documentation-boundary-audit.mjs +0 -305
- package/scripts/explain-on-demand-audit.mjs +0 -427
- package/scripts/forbidden-content-check.mjs +0 -123
- package/scripts/frontend-usability-audit.mjs +0 -243
- package/scripts/generate-sbom.mjs +0 -61
- package/scripts/governance-weekly-report.mjs +0 -466
- package/scripts/llm-judge/checklist-loader.mjs +0 -45
- package/scripts/llm-judge/constants.mjs +0 -66
- package/scripts/llm-judge/diff-collection.mjs +0 -74
- package/scripts/llm-judge/prompting.mjs +0 -78
- package/scripts/llm-judge/providers.mjs +0 -111
- package/scripts/llm-judge/verdict.mjs +0 -134
- package/scripts/llm-judge.mjs +0 -200
- package/scripts/memory-continuity-benchmark.mjs +0 -322
- package/scripts/quality-trend-report.mjs +0 -289
- package/scripts/release-gate/audit-checks.mjs +0 -441
- package/scripts/release-gate/constants.mjs +0 -53
- package/scripts/release-gate/runtime.mjs +0 -63
- package/scripts/release-gate/static-checks.mjs +0 -182
- package/scripts/release-gate.mjs +0 -42
- package/scripts/rules-guardian-audit.mjs +0 -580
- package/scripts/single-source-lazy-loading-audit.mjs +0 -449
- package/scripts/sync-thin-adapters.mjs +0 -72
- package/scripts/token-optimization-benchmark.mjs +0 -252
- package/scripts/ui-design-judge/constants.mjs +0 -13
- package/scripts/ui-design-judge/design-execution-summary.mjs +0 -81
- package/scripts/ui-design-judge/git-input.mjs +0 -134
- package/scripts/ui-design-judge/prompting.mjs +0 -52
- package/scripts/ui-design-judge/providers.mjs +0 -102
- package/scripts/ui-design-judge/reporting.mjs +0 -183
- package/scripts/ui-design-judge/rubric-calibration.mjs +0 -214
- package/scripts/ui-design-judge/rubric-goldset.json +0 -188
- package/scripts/ui-design-judge.mjs +0 -319
- package/scripts/ui-rubric-calibration.mjs +0 -35
- package/scripts/validate/config.mjs +0 -519
- package/scripts/validate/coverage-checks.mjs +0 -446
- package/scripts/validate/file-structure.mjs +0 -162
- package/scripts/validate/markdown-content.mjs +0 -109
- package/scripts/validate/project-metadata.mjs +0 -166
- package/scripts/validate.mjs +0 -338
|
@@ -1,31 +0,0 @@
|
|
|
1
|
-
# Prompt: Review Code
|
|
2
|
-
|
|
3
|
-
Use this when reviewing current changes, a pull request, or a focused file set.
|
|
4
|
-
|
|
5
|
-
```text
|
|
6
|
-
Review the code with a production-risk mindset.
|
|
7
|
-
|
|
8
|
-
Before reviewing:
|
|
9
|
-
1. Read AGENTS.md.
|
|
10
|
-
2. Read .agent-context/review-checklists/pr-checklist.md.
|
|
11
|
-
3. Read .agent-context/review-checklists/architecture-review.md only when architecture or boundaries changed.
|
|
12
|
-
4. Load only the rules relevant to the changed scope.
|
|
13
|
-
5. For UI changes, load .agent-context/prompts/bootstrap-design.md, .agent-context/rules/frontend-architecture.md, and docs/DESIGN.md when present.
|
|
14
|
-
6. Enforce Universal SOP hard gate: block coding flow when required project docs are missing (root `README.md`; `docs/doc-index.md` when `docs/` exists; `docs/project-brief.md`; `docs/architecture-decision-record.md`; `docs/flow-overview.md`; `docs/database-schema.md` when persistent data exists; `docs/api-contract.md` when API or web application flows exist; and for UI scope `docs/DESIGN.md`).
|
|
15
|
-
7. Enforce single-source and lazy-loading policy: canonical rule source must be explicitly enforced, global domain governance must load lazily based on touched scope, and conflicting duplicate rule instructions must not appear during normal flow.
|
|
16
|
-
|
|
17
|
-
Prioritize findings in this order:
|
|
18
|
-
1. Correctness, data loss, security, privacy, auth, and permission risks.
|
|
19
|
-
2. Public contract drift: APIs, events, CLI behavior, data model, UI contract, docs.
|
|
20
|
-
3. Missing tests for changed behavior.
|
|
21
|
-
4. Architecture boundary drift and maintainability risk.
|
|
22
|
-
5. Performance and accessibility issues with concrete impact.
|
|
23
|
-
|
|
24
|
-
For every finding:
|
|
25
|
-
- include file and line
|
|
26
|
-
- explain the real risk
|
|
27
|
-
- reference the rule or contract only when it materially supports the finding
|
|
28
|
-
- propose the smallest safe fix
|
|
29
|
-
|
|
30
|
-
Do not create stack-specific governance concerns. Use project framework details only for concrete implementation risks proven by changed code, docs, or current official documentation.
|
|
31
|
-
```
|
|
@@ -1,57 +0,0 @@
|
|
|
1
|
-
# Architecture Review Checklist
|
|
2
|
-
|
|
3
|
-
Use this when module structure, feature shape, public contracts, topology, or refactoring are in scope.
|
|
4
|
-
|
|
5
|
-
## Boundary Review
|
|
6
|
-
|
|
7
|
-
- [ ] The changed code has clear transport, application, domain, and infrastructure boundaries where those layers exist.
|
|
8
|
-
- [ ] Business policy is not hidden in transport handlers, UI adapters, database queries, framework glue, or generated code.
|
|
9
|
-
- [ ] Controllers and route handlers only translate protocol input/output, enforce edge checks, and delegate business flow.
|
|
10
|
-
- [ ] Service or use-case code owns orchestration, transactions, state transitions, and idempotency decisions.
|
|
11
|
-
- [ ] Repository or adapter code owns persistence and external IO details without hiding product policy in queries.
|
|
12
|
-
- [ ] Global backend/API governance is used directly; no stack-specific adapter or framework-specific rule fork was introduced.
|
|
13
|
-
- [ ] Internal models do not leak across public API, event, CLI, library, or UI contracts without a deliberate mapping.
|
|
14
|
-
- [ ] Modules import through public entrypoints instead of deep internal paths.
|
|
15
|
-
- [ ] Circular dependencies are absent or explicitly removed.
|
|
16
|
-
|
|
17
|
-
## Structure Review
|
|
18
|
-
|
|
19
|
-
## Backend Universal Principles
|
|
20
|
-
|
|
21
|
-
- [ ] No clever hacks in backend and shared core modules
|
|
22
|
-
- [ ] No premature abstraction (base classes/util layers created only after repeated stable patterns)
|
|
23
|
-
- [ ] Readability over brevity for maintainability
|
|
24
|
-
- [ ] The project structure follows existing repo conventions unless a change was approved.
|
|
25
|
-
- [ ] New structure is feature/domain-oriented when that improves discoverability.
|
|
26
|
-
- [ ] Large files or modules are split around real responsibilities, not arbitrary layers.
|
|
27
|
-
- [ ] Shared code is genuinely shared and not domain-specific behavior disguised as utility.
|
|
28
|
-
- [ ] Names describe product meaning, not generic plumbing.
|
|
29
|
-
|
|
30
|
-
## Topology Review
|
|
31
|
-
|
|
32
|
-
- [ ] Monolith remains acceptable when boundaries, team size, consistency needs, and operations favor one deployable.
|
|
33
|
-
- [ ] Service extraction is backed by current evidence: ownership, scaling, compliance, fault isolation, or deploy cadence.
|
|
34
|
-
- [ ] Service boundaries define ownership, contract, data boundary, observability, timeouts, retry, and recovery behavior.
|
|
35
|
-
- [ ] Event or realtime architecture is justified by product need, not trend pressure.
|
|
36
|
-
|
|
37
|
-
## Contract Review
|
|
38
|
-
|
|
39
|
-
- [ ] API, event, CLI, library, data, and UI contracts are documented before or alongside implementation.
|
|
40
|
-
- [ ] Schema and validation strategy matches the project’s chosen runtime and official docs.
|
|
41
|
-
- [ ] Error contracts are safe, stable, and do not leak internals.
|
|
42
|
-
- [ ] List endpoints have bounded pagination, limits, or streaming behavior documented.
|
|
43
|
-
- [ ] Sensitive mutations have documented idempotency, retry, and duplicate-submit behavior.
|
|
44
|
-
- [ ] Error contracts document stable codes, safe trace or correlation identifiers, and any Problem Details-style fields when exposed.
|
|
45
|
-
- [ ] Async/event contracts document retry, ordering, duplicate handling, and dead-letter or recovery behavior.
|
|
46
|
-
- [ ] Migration and rollback plans exist for risky data or public contract changes.
|
|
47
|
-
|
|
48
|
-
## Operational Review
|
|
49
|
-
|
|
50
|
-
- [ ] Critical flows have tests at the right level.
|
|
51
|
-
- [ ] Observability, logging, and health checks match the project’s runtime and risk level.
|
|
52
|
-
- [ ] Security assumptions are documented and enforced at trust boundaries.
|
|
53
|
-
- [ ] Performance-sensitive paths avoid avoidable repeated work, unbounded lists, and hidden blocking operations.
|
|
54
|
-
- [ ] Relational reads avoid N+1 patterns or include an explicit query-shape rationale.
|
|
55
|
-
- [ ] Multi-table or cross-resource writes are transactional or include a documented compensating recovery path.
|
|
56
|
-
- [ ] Dual-write database plus message flows use an outbox or equivalent atomicity and replay strategy.
|
|
57
|
-
- [ ] Cross-service consistency avoids default two-phase commit and defines saga, compensation, or recovery behavior when needed.
|
|
@@ -1,136 +0,0 @@
|
|
|
1
|
-
# PR Checklist - Quality Gate
|
|
2
|
-
|
|
3
|
-
Run this before declaring a task done. Apply only the sections relevant to the changed scope, but do not skip correctness, security, testing, or docs checks when the change touches them.
|
|
4
|
-
|
|
5
|
-
## 1. Repo Context
|
|
6
|
-
|
|
7
|
-
- [ ] The agent read `AGENTS.md` and the smallest relevant rule set.
|
|
8
|
-
- [ ] For non-trivial coding, review, planning, or governance work, the agent produced a Bootstrap Receipt with loaded files, selected rules, skipped rules, unreachable files, and validation plan before implementation output.
|
|
9
|
-
- [ ] Risky actions used the AGENTS.md Bounded Reflection block with valid rule IDs, one-line rationale, and no copied rule prose or hidden chain-of-thought.
|
|
10
|
-
- [ ] Existing project context came from real files, docs, package metadata, and changed code, not folder name alone.
|
|
11
|
-
- [ ] Runtime, framework, library, topology, and design choices are explicit user constraints or agent recommendations from current evidence.
|
|
12
|
-
- [ ] No offline default stack, blueprint, vendor, or visual style was treated as authoritative.
|
|
13
|
-
|
|
14
|
-
## 2. Correctness
|
|
15
|
-
|
|
16
|
-
- [ ] The changed behavior matches the user request.
|
|
17
|
-
- [ ] Existing behavior is preserved unless the user approved a behavior change.
|
|
18
|
-
- [ ] Edge cases, empty states, error paths, and rollback/recovery paths are handled.
|
|
19
|
-
- [ ] Public contracts remain stable or are versioned and documented.
|
|
20
|
-
|
|
21
|
-
## 3. Architecture
|
|
22
|
-
|
|
23
|
-
### 2. Architecture (→ rules/architecture.md)
|
|
24
|
-
|
|
25
|
-
- [ ] Layer and module boundaries are clear for the project’s chosen structure.
|
|
26
|
-
- [ ] No clever hacks in backend and shared core modules
|
|
27
|
-
- [ ] No premature abstraction (base classes/util layers created only after repeated stable patterns)
|
|
28
|
-
- [ ] Readability over brevity for maintainability
|
|
29
|
-
- [ ] Complexity budget was applied: equivalent behavior uses fewer moving parts without losing validation, error handling, fallbacks, accessibility, tests, or security boundaries.
|
|
30
|
-
- [ ] Natural implementation pass was applied: the main flow is traceable, names are domain-specific, helpers carry real meaning, and compact code did not hide safeguards.
|
|
31
|
-
- [ ] Controllers, route handlers, and transport adapters do not contain business policy, raw queries, or cross-resource orchestration.
|
|
32
|
-
- [ ] Services or use cases own business flow, transaction boundaries, and mutation safety.
|
|
33
|
-
- [ ] Repositories or adapters own persistence/external IO details without hiding business decisions.
|
|
34
|
-
- [ ] Backend/API governance was applied through global domain rules, not stack-specific adapters or framework-only branches.
|
|
35
|
-
- [ ] Code is grouped by feature/domain where that improves maintainability.
|
|
36
|
-
- [ ] Cross-module access uses public contracts instead of internal file reach-through.
|
|
37
|
-
- [ ] Files above roughly 1000 lines were split or explicitly justified.
|
|
38
|
-
- [ ] Monolith/service split decisions are evidence-backed, not fashion-driven.
|
|
39
|
-
|
|
40
|
-
## 4. Security And Privacy
|
|
41
|
-
|
|
42
|
-
- [ ] External input is validated at trust boundaries using the project’s chosen validation approach.
|
|
43
|
-
- [ ] Body, query, params, headers, cookies, uploads, webhooks, and job payloads are treated as untrusted until validated and normalized.
|
|
44
|
-
- [ ] Secrets, tokens, credentials, and private data are not committed or logged.
|
|
45
|
-
- [ ] Authorization is enforced at a trusted boundary.
|
|
46
|
-
- [ ] Error responses and logs do not leak internals.
|
|
47
|
-
- [ ] Least privilege, resource-level authorization, and secret handling are preserved where sensitive data or privileged actions are touched.
|
|
48
|
-
- [ ] Dependency or platform security claims are based on current official docs or repo evidence.
|
|
49
|
-
|
|
50
|
-
## 5. Testing
|
|
51
|
-
|
|
52
|
-
- [ ] Changed behavior has appropriate tests at the smallest useful level.
|
|
53
|
-
- [ ] API changes cover validation, authorization, documented error shape, pagination defaults, and empty states where relevant.
|
|
54
|
-
- [ ] Sensitive mutations include idempotency or duplicate-submit coverage where duplicate side effects would be harmful.
|
|
55
|
-
- [ ] Tests assert behavior and contracts, not implementation trivia.
|
|
56
|
-
- [ ] Critical flows include failure-path coverage.
|
|
57
|
-
- [ ] Test fixtures are readable and do not hide the behavior under test.
|
|
58
|
-
|
|
59
|
-
## 6. Docs And Contracts
|
|
60
|
-
|
|
61
|
-
### 10. Documentation
|
|
62
|
-
|
|
63
|
-
- [ ] Scope applied: This applies to documentation, release notes, onboarding text, review summaries, and agent-facing explanations
|
|
64
|
-
- [ ] Style scope review is advisory and does not block merge when API docs are synced in the same commit and contract details are correct
|
|
65
|
-
- [ ] Required docs exist before implementation: public and developer root README; project brief; architecture decision; flow overview; API/public contract when relevant; data model when relevant; and UI design contract when relevant.
|
|
66
|
-
- [ ] `docs/doc-index.md` exists whenever `docs/` exists and acts as a compact read-routing map instead of duplicating requirements or architecture.
|
|
67
|
-
- [ ] For docs-only or docs-first requests, implementation code was not changed unless the user explicitly asked for it or approved an implementation plan.
|
|
68
|
-
- [ ] Formal project docs use English by default unless the user requested another language or existing docs established one.
|
|
69
|
-
- [ ] Docs cover feature plan, architecture rationale, public contracts, data model, UI/design, security assumptions, testing strategy, delivery flow, and next validation actions where relevant.
|
|
70
|
-
- [ ] API, event, CLI, library, data, and UI contract changes update docs in the same scope.
|
|
71
|
-
- [ ] Public surface changes fail review if documentation updates are missing or stale in the same scope
|
|
72
|
-
- [ ] Documentation checks stay boundary-aware and only enforce touched scopes
|
|
73
|
-
- [ ] Facts, assumptions, and next actions are separated when context is incomplete.
|
|
74
|
-
- [ ] No emoji in formal documentation or review summaries
|
|
75
|
-
- [ ] Documentation uses plain English and avoids AI cliches
|
|
76
|
-
- [ ] Root README is public and developer friendly, even for private projects, and does not contain secrets, internal agent notes, private reasoning, or governance policy dumps.
|
|
77
|
-
- [ ] Documentation grows with the project: README and matching docs were updated when setup, runtime, architecture, public contracts, data shape, deployment, validation, or UI scope changed.
|
|
78
|
-
- [ ] Documentation file count stayed intentional: new docs files were added only for stable, distinct, or long workflows.
|
|
79
|
-
- [ ] PRD, SRS, technical-design, or separate ERD files were added only when project evidence justified a distinct document.
|
|
80
|
-
|
|
81
|
-
## 7. UI And Accessibility
|
|
82
|
-
|
|
83
|
-
- [ ] UI work follows `docs/DESIGN.md`.
|
|
84
|
-
- [ ] Visual direction is project-specific and not a template/default component-kit habit.
|
|
85
|
-
- [ ] UI work includes a Motion/Palette Decision, and product categories were treated as heuristics rather than style presets.
|
|
86
|
-
- [ ] Responsive behavior recomposes content and priority, not only shrinking desktop layout.
|
|
87
|
-
- [ ] Accessibility hard requirements are preserved: keyboard access, focus visibility, contrast, target size, status feedback, and no color-only meaning.
|
|
88
|
-
- [ ] Motion is treated as part of the design language for modern UI work, with reduced-motion and performance safeguards instead of defaulting to static screens.
|
|
89
|
-
- [ ] Broad redesigns pass the old-design regression test: the result is not the previous composition with animation, depth, media, or interaction density removed.
|
|
90
|
-
- [ ] UI work records an agent-chosen ambition level; broad screens and redesigns researched an expressive path first, and any downshift names a concrete blocker plus replacement interaction quality.
|
|
91
|
-
- [ ] UI foundation choices are dynamic and product-fit; no shadcn, native-only, Tailwind-only, or component-kit default was selected by habit or avoided from dependency fear.
|
|
92
|
-
- [ ] Design intent separates locked outcomes from flexible expression; candidate signature moves, exact token primitives, literal anchor artifacts, and component-kit skins were not treated as permanent requirements without evidence or user approval.
|
|
93
|
-
|
|
94
|
-
## 8. Dependencies And Runtime
|
|
95
|
-
|
|
96
|
-
- [ ] New dependencies are justified by capability, maintenance health, bundle/runtime cost, and current official docs.
|
|
97
|
-
- [ ] Dependency avoidance was not treated as a default virtue; lightweight maintained libraries were considered when they improve correctness, accessibility, UX, maintainability, or delivery speed.
|
|
98
|
-
- [ ] Official setup flows are preferred when they produce better-supported current defaults.
|
|
99
|
-
- [ ] Docker, framework, package, and ecosystem claims were checked live when they could be stale.
|
|
100
|
-
- [ ] Token optimization and memory continuity defaults remain enabled unless the user explicitly opts out.
|
|
101
|
-
- [ ] If `.agent-context/state/active-memory.json` exists and the task made material project progress, the agent refreshed it directly before completion and preserved privacy/user-owned entries.
|
|
102
|
-
|
|
103
|
-
## 9. State And Governance
|
|
104
|
-
|
|
105
|
-
### 11. Context-Triggered Audit Mode
|
|
106
|
-
|
|
107
|
-
- [ ] Strict audit mode activates automatically on review and PR-intent workflows
|
|
108
|
-
- [ ] Small edits avoid heavy checks by default unless strict mode is explicitly requested
|
|
109
|
-
- [ ] User can always force strict audit mode manually
|
|
110
|
-
- [ ] Session handoff includes active architecture contract summary.
|
|
111
|
-
- [ ] Drift detection warns before direction changes.
|
|
112
|
-
- [ ] Direction changes require explicit user confirmation.
|
|
113
|
-
- [ ] Default responses avoid unnecessary state-file internals.
|
|
114
|
-
- [ ] State internals are exposed only on explicit request.
|
|
115
|
-
- [ ] Diagnostic mode can explain relevant state decisions when needed.
|
|
116
|
-
- [ ] Canonical rule source is explicitly defined and enforced
|
|
117
|
-
- [ ] Global domain governance is loaded lazily based on touched scope
|
|
118
|
-
- [ ] No conflicting duplicate rule instructions during normal flow
|
|
119
|
-
- [ ] Canonical rule source is explicit and duplicate/conflicting instructions are avoided.
|
|
120
|
-
|
|
121
|
-
### 15. Universal SOP Consolidation
|
|
122
|
-
|
|
123
|
-
- [ ] `.agent-context/rules/` remains the default guidance source for implementation and review.
|
|
124
|
-
- [ ] Security and testing requirements remain mandatory after static template purge.
|
|
125
|
-
- [ ] Coding flow is blocked if `docs/project-brief.md` is missing
|
|
126
|
-
- [ ] Coding flow is blocked if `docs/architecture-decision-record.md` (or `docs/Architecture-Decision-Record.md`) is missing
|
|
127
|
-
- [ ] Coding flow is blocked if root `README.md` is missing
|
|
128
|
-
- [ ] Coding flow is blocked if `docs/doc-index.md` is missing while `docs/` exists
|
|
129
|
-
- [ ] Coding flow is blocked if `docs/flow-overview.md` is missing
|
|
130
|
-
- [ ] Coding flow is blocked if `docs/database-schema.md` is missing while the project uses persistent data
|
|
131
|
-
- [ ] Coding flow is blocked if `docs/api-contract.md` is missing while the project exposes API or web application flows
|
|
132
|
-
- [ ] UI implementation flow is blocked if `docs/DESIGN.md` is missing
|
|
133
|
-
|
|
134
|
-
## Verdict
|
|
135
|
-
|
|
136
|
-
Report findings first, ordered by severity, with file/line references and concrete fixes. If no findings exist, say that explicitly and name any residual risk.
|
|
@@ -1,24 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: API
|
|
3
|
-
domain: api-docs
|
|
4
|
-
priority: high
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [api-docs, api, contract, documentation]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# API Contract Boundary
|
|
11
|
-
|
|
12
|
-
## API-001: Execution Rules
|
|
13
|
-
1. Sync docs in the same commit when changing API, CLI, or schema.
|
|
14
|
-
2. Root README.md is mandatory. Keep it overview-level. No secrets.
|
|
15
|
-
3. Keep `docs/doc-index.md` as the routing map.
|
|
16
|
-
4. Use OpenAPI 3.1 for HTTP APIs by default.
|
|
17
|
-
5. Idempotency is mandatory for side-effect mutations (POST/PUT/PATCH).
|
|
18
|
-
6. Use Mermaid.js natively for all diagrams. No PlantUML/D2.
|
|
19
|
-
|
|
20
|
-
## API-002: Human Writing & Anti-Slop (Mandatory)
|
|
21
|
-
1. NO EMOJI in any formal documentation, code, or review summaries. This is absolute.
|
|
22
|
-
2. Use plain English and avoid generic AI cliches ("AI slop") in generated docs.
|
|
23
|
-
3. Keep code clean and efficient. Do not write convoluted/long code if a shorter, efficient solution exists without losing functionality.
|
|
24
|
-
4. Keep documentation short, clear, and directly actionable.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: VER
|
|
3
|
-
domain: api-versioning
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [api-versioning, versioning, deprecation]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# API Versioning Boundary
|
|
11
|
-
|
|
12
|
-
## VER-001: Execution Rules
|
|
13
|
-
1. Never introduce breaking changes without versioning.
|
|
14
|
-
2. Maintain backward compatibility.
|
|
15
|
-
3. Use header-based or URL-path versioning explicitly.
|
|
16
|
-
4. Document deprecation windows before sunsetting endpoints.
|
|
@@ -1,33 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: ARCH
|
|
3
|
-
domain: architecture
|
|
4
|
-
priority: critical
|
|
5
|
-
scope: all-tasks
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [architecture, arch, boundary, system]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Architecture Boundary
|
|
11
|
-
|
|
12
|
-
## ARCH-001: Execution Rules
|
|
13
|
-
1. Rely on ESLint/Linters for structural enforcement.
|
|
14
|
-
2. Require README.md and docs/doc-index.md.
|
|
15
|
-
3. Run `npm run validate` to enforce architecture invariants.
|
|
16
|
-
4. Do not invent custom crypto, custom state management, or custom routing. Use standard libraries.
|
|
17
|
-
|
|
18
|
-
## ARCH-005: Rules as Guardian (Cross-Session Consistency)
|
|
19
|
-
1. Session handoff must include active architecture contract summary.
|
|
20
|
-
2. Detect drift before changing runtime choices, topology, public contracts, or core patterns.
|
|
21
|
-
3. Direction changes require explicit user confirmation before applying changes.
|
|
22
|
-
|
|
23
|
-
## ARCH-006: Invisible State Management with Explain-on-Demand
|
|
24
|
-
1. Default responses must avoid unnecessary state-file internals.
|
|
25
|
-
2. State internals are exposed only on explicit user request.
|
|
26
|
-
3. Diagnostic mode explains relevant state decisions when needed.
|
|
27
|
-
|
|
28
|
-
## ARCH-007: Single Source of Truth and Lazy Rule Loading
|
|
29
|
-
1. Canonical rule source is AGENTS.md.
|
|
30
|
-
2. Load global domain rules lazily based on touched scope.
|
|
31
|
-
3. Do not create or load stack-specific governance adapters as the baseline.
|
|
32
|
-
|
|
33
|
-
|
|
@@ -1,17 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: JOB
|
|
3
|
-
domain: background-jobs
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [background-jobs, worker, cron, queue]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Background Jobs Boundary
|
|
11
|
-
|
|
12
|
-
## JOB-001: Execution Rules
|
|
13
|
-
1. All jobs must be idempotent.
|
|
14
|
-
2. Offload heavy processing (>500ms) to background queues.
|
|
15
|
-
3. Use dead-letter queues (DLQ) for failed jobs.
|
|
16
|
-
4. Emit telemetry on job failure/success.
|
|
17
|
-
5. Jobs must have timeouts and retry limits.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: CFG
|
|
3
|
-
domain: config-and-flags
|
|
4
|
-
priority: high
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [configuration, feature-flags, environment, secrets]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Configuration Boundary
|
|
11
|
-
|
|
12
|
-
## CFG-001: Execution Rules
|
|
13
|
-
1. Inject secrets via environment variables.
|
|
14
|
-
2. Never store secrets in code. Use .env.example for templates.
|
|
15
|
-
3. Use feature flags for incremental rollouts.
|
|
16
|
-
4. Validate config shapes at startup. Fail fast if invalid.
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: DATA
|
|
3
|
-
domain: database-design
|
|
4
|
-
priority: high
|
|
5
|
-
scope: data
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [database-design, data, schema]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Data Design Boundary
|
|
11
|
-
|
|
12
|
-
## DATA-001: Execution Rules
|
|
13
|
-
1. Avoid N+1 queries. Use eager loading or batching.
|
|
14
|
-
2. Paginate growable datasets.
|
|
15
|
-
3. Multi-table mutations must run inside transactions.
|
|
16
|
-
4. Monetary amounts must be integer (minor units) or exact decimal. NO floats.
|
|
17
|
-
5. Store real-world timestamps in UTC. NO naive timestamps.
|
|
18
|
-
6. Use optimistic concurrency tokens (ETag/version) for shared mutable resources.
|
|
@@ -1,17 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: DOCK
|
|
3
|
-
domain: docker-runtime
|
|
4
|
-
priority: high
|
|
5
|
-
scope: infra
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [docker-runtime, docker, runtime]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Docker Runtime Boundary
|
|
11
|
-
|
|
12
|
-
## DOCK-001: Execution Rules
|
|
13
|
-
1. ALWAYS perform Live Research / Web Search for latest official Docker documentation before writing configuration.
|
|
14
|
-
2. Keep Dev and Prod stages separate.
|
|
15
|
-
3. Use minimal trusted base images. Avoid running as root in Prod.
|
|
16
|
-
4. Never bake secrets into image layers.
|
|
17
|
-
5. Define explicit healthchecks in production compose files.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: ERR
|
|
3
|
-
domain: error-handling
|
|
4
|
-
priority: high
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [error-handling, errors, exception]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Error Handling Boundary
|
|
11
|
-
|
|
12
|
-
## ERR-001: Execution Rules
|
|
13
|
-
1. Fail fast on invalid input.
|
|
14
|
-
2. Do not leak stack traces to clients. Log them securely.
|
|
15
|
-
3. Use standardized machine-readable error codes (RFC 9457).
|
|
16
|
-
4. Distinguish client errors (4xx) from server errors (5xx).
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: EVT
|
|
3
|
-
domain: event-driven
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [event-driven, events, kafka, rabbitmq]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Event Driven Boundary
|
|
11
|
-
|
|
12
|
-
## EVT-001: Execution Rules
|
|
13
|
-
1. Events must be immutable.
|
|
14
|
-
2. Consumers must be idempotent.
|
|
15
|
-
3. Handle out-of-order events securely.
|
|
16
|
-
4. Always implement Dead Letter Queues (DLQ) for poison messages.
|
|
@@ -1,54 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: FE
|
|
3
|
-
domain: frontend-architecture
|
|
4
|
-
priority: high
|
|
5
|
-
scope: ui
|
|
6
|
-
last_validated: 2026-05-17
|
|
7
|
-
applies_to: [frontend, fullstack]
|
|
8
|
-
keywords: [frontend-architecture, fe, ui, engineering]
|
|
9
|
-
---
|
|
10
|
-
|
|
11
|
-
# Frontend Engineering Invariants (Tier 2)
|
|
12
|
-
|
|
13
|
-
Load this rule for UI-facing engineering work. This file contains strict engineering constraints.
|
|
14
|
-
|
|
15
|
-
## FE-009: Accessibility
|
|
16
|
-
1. WCAG 2.2 AA is the hard floor. APCA is advisory perceptual tuning only.
|
|
17
|
-
2. Hard checks: focus visibility, target size, keyboard access, color-only meaning.
|
|
18
|
-
|
|
19
|
-
## FE-010: CSS Production Hardening
|
|
20
|
-
1. Plan overflow, wrapping, truncation, empty, loading, and error behavior.
|
|
21
|
-
2. Prefer logical composition primitives (container queries, subgrid).
|
|
22
|
-
|
|
23
|
-
## FE-011: Implementation Boundaries
|
|
24
|
-
1. Follow the shipped project stack and current repo patterns.
|
|
25
|
-
2. Do not hardcode framework-specific state management or data fetching libraries, or rigid component doctrines.
|
|
26
|
-
|
|
27
|
-
## FE-012: Data State Surface
|
|
28
|
-
1. Explicitly handle empty, loading, error, and offline states.
|
|
29
|
-
2. Reject "spinner everywhere".
|
|
30
|
-
|
|
31
|
-
## FE-013: Background and Wallpaper Discipline
|
|
32
|
-
1. Decorative geometry are invalid as wallpaper.
|
|
33
|
-
2. Use grids/lines only for specific functional roles (alignment, map, crop).
|
|
34
|
-
|
|
35
|
-
## FE-014: Production Content Policy
|
|
36
|
-
1. No visible testing, demo, sample, placeholder, lorem, or TODO labels in production UI.
|
|
37
|
-
|
|
38
|
-
## FE-015: Motion Implementation Budget
|
|
39
|
-
1. Keep reduced-motion, keyboard, loading, performance, and mobile fallbacks explicit.
|
|
40
|
-
|
|
41
|
-
## FE-016: Library and Design-Intent Discipline
|
|
42
|
-
1. Use component kits or headless primitives for behavior and accessibility when they fit.
|
|
43
|
-
|
|
44
|
-
## FE-017: Interactivity Priority
|
|
45
|
-
1. The interactive boundary must be drawn deliberately. Keep client-side state boundaries small.
|
|
46
|
-
2. Measure responsiveness through Interaction to Next Paint (INP).
|
|
47
|
-
|
|
48
|
-
## FE-018: Internationalization as Layout
|
|
49
|
-
1. Direction-sensitive spacing, alignment, and positioning must use CSS logical properties (e.g., `margin-inline-start`).
|
|
50
|
-
2. Plan a documented text-expansion budget for target locales.
|
|
51
|
-
|
|
52
|
-
## FE-019: Theme as Context
|
|
53
|
-
1. A theme switch is a change in lighting and surface model, not a color inversion. Re-derive tokens per theme.
|
|
54
|
-
2. Brand colors carried across themes must be individually verified against the active theme's contrast floor.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: SVC
|
|
3
|
-
domain: microservices
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [microservices, monolith]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Microservices Boundary
|
|
11
|
-
|
|
12
|
-
## SVC-001: Execution Rules
|
|
13
|
-
1. Default to modular monoliths unless scale explicitly dictates microservices.
|
|
14
|
-
2. Independent services must own their data. NO shared databases.
|
|
15
|
-
3. Cross-service calls must have timeouts and retries.
|
|
16
|
-
4. Prefer async choreographies over distributed two-phase commits.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: MIG
|
|
3
|
-
domain: migrations
|
|
4
|
-
priority: high
|
|
5
|
-
scope: data
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [migrations, schema, database]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Migrations Boundary
|
|
11
|
-
|
|
12
|
-
## MIG-001: Execution Rules
|
|
13
|
-
1. Schema changes MUST have a versioned migration.
|
|
14
|
-
2. Migrations must be reversible (down-migrations) or have a recovery plan.
|
|
15
|
-
3. Never modify past migrations once merged. Create a new one.
|
|
16
|
-
4. Use concurrent index builds in production.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: OBS
|
|
3
|
-
domain: observability
|
|
4
|
-
priority: high
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [observability, telemetry, logs, metrics]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Observability Boundary
|
|
11
|
-
|
|
12
|
-
## OBS-001: Execution Rules
|
|
13
|
-
1. Emit telemetry for failures, degraded states, and security events.
|
|
14
|
-
2. Inject correlation IDs on inbound requests and propagate them downstream.
|
|
15
|
-
3. Do not log PII, secrets, or credentials.
|
|
16
|
-
4. Measure latency, traffic, errors, and saturation.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: PERF
|
|
3
|
-
domain: performance
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [performance, perf, optimization]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Performance Boundary
|
|
11
|
-
|
|
12
|
-
## PERF-001: Execution Rules
|
|
13
|
-
1. Optimize DB queries before adding caches.
|
|
14
|
-
2. Compress network payloads.
|
|
15
|
-
3. Bound memory usage (paginate, stream). No unbounded arrays.
|
|
16
|
-
4. Cache aggressively but define explicit cache invalidation.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: RT
|
|
3
|
-
domain: realtime
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [realtime, websockets, sse]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Realtime Boundary
|
|
11
|
-
|
|
12
|
-
## RT-001: Execution Rules
|
|
13
|
-
1. Use SSE for one-way server-to-client streams. Use WebSockets only for true bi-directional.
|
|
14
|
-
2. Realtime connections must gracefully degrade to polling if blocked.
|
|
15
|
-
3. Implement ping/pong heartbeats to drop dead connections.
|
|
16
|
-
4. Scale pub/sub via Redis or dedicated brokers.
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: RES
|
|
3
|
-
domain: resilience
|
|
4
|
-
priority: critical
|
|
5
|
-
scope: backend
|
|
6
|
-
applies_to: [backend, fullstack]
|
|
7
|
-
keywords: [resilience, timeout, retry]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Resilience Boundary
|
|
11
|
-
|
|
12
|
-
## RES-001: Execution Rules
|
|
13
|
-
1. Every outbound network call MUST have a strict timeout.
|
|
14
|
-
2. Retries MUST use exponential backoff with jitter and max attempt limits.
|
|
15
|
-
3. Only retry idempotent operations.
|
|
16
|
-
4. Fail fast using Circuit Breakers / Load Shedding on unhealthy dependencies.
|
|
17
|
-
5. Provide graceful degradation on non-critical dependency failures.
|
|
18
|
-
6. Enforce backpressure on bounded consumers.
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: SEC
|
|
3
|
-
domain: security
|
|
4
|
-
priority: critical
|
|
5
|
-
scope: all-tasks
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [security, boundary, zero-trust]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Security Boundary
|
|
11
|
-
|
|
12
|
-
## SEC-001: Execution Rules
|
|
13
|
-
1. Validate and normalize ALL inputs.
|
|
14
|
-
2. Parameterize all SQL queries. Never interpolate input into DB or Shell.
|
|
15
|
-
3. Hash passwords with Argon2/Bcrypt. NEVER store plain text or use MD5/SHA.
|
|
16
|
-
4. NEVER commit secrets, tokens, or credentials.
|
|
17
|
-
5. Implement resource-aware authorization (Row-Level), not just Authentication.
|
|
18
|
-
6. Verify service-to-service cryptographic identity.
|
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
id_prefix: TEST
|
|
3
|
-
domain: testing
|
|
4
|
-
priority: medium
|
|
5
|
-
scope: all-tasks
|
|
6
|
-
applies_to: [backend, frontend, fullstack]
|
|
7
|
-
keywords: [testing, tests, coverage]
|
|
8
|
-
---
|
|
9
|
-
|
|
10
|
-
# Testing Boundary
|
|
11
|
-
|
|
12
|
-
## TEST-001: Execution Rules
|
|
13
|
-
1. Write tests for business logic and boundary failures.
|
|
14
|
-
2. Test error states, not just the happy path.
|
|
15
|
-
3. Use deterministic mocks for external services.
|
|
16
|
-
4. Ensure CI pipelines block on test failures.
|
|
@@ -1,25 +0,0 @@
|
|
|
1
|
-
# State Artifacts
|
|
2
|
-
|
|
3
|
-
Use this directory as a typed governance state surface, not as a dump folder.
|
|
4
|
-
|
|
5
|
-
Tracked seed/config artifacts:
|
|
6
|
-
- `architecture-map.md`
|
|
7
|
-
- `dependency-map.md`
|
|
8
|
-
- `benchmark-comparison-schema.json`
|
|
9
|
-
- `benchmark-reproducibility.json`
|
|
10
|
-
- `benchmark-thresholds.json`
|
|
11
|
-
- `benchmark-writer-judge-config.json`
|
|
12
|
-
- `memory-adapter-contract.json`
|
|
13
|
-
- `memory-schema-v1.json`
|
|
14
|
-
|
|
15
|
-
Tracked operational artifact:
|
|
16
|
-
- `onboarding-report.json` stays tracked in this repository because internal audits read it as the current repository onboarding state. In installed projects, `init` and `upgrade` regenerate this file.
|
|
17
|
-
|
|
18
|
-
Local-only/generated artifacts:
|
|
19
|
-
- `active-memory.json`
|
|
20
|
-
- `llm-judge-report.json`
|
|
21
|
-
- `token-saver/`
|
|
22
|
-
- `token-optimization-report.json`
|
|
23
|
-
- benchmark, trend, weekly governance, and quality report outputs
|
|
24
|
-
|
|
25
|
-
Do not treat generated reports or archived research as current project truth. Rerun the matching `npm run benchmark:*`, `npm run report:*`, or `npm run audit:*` command when fresh evidence is needed.
|