Commit Graph

3383 Commits

Author SHA1 Message Date
diegosouzapw
d28da844c1 merge(F11): fix 4 review gaps (raw text/markdown, README, dead code, generator-output) 2026-05-28 02:33:53 -03:00
diegosouzapw
431d31a713 chore(coverage): lower coverage gate to 40/40/40/40 (statements/lines/functions/branches) 2026-05-28 02:33:53 -03:00
diegosouzapw
1290f3a4c1 docs(orchestration): create 15-generator-output.md (GAP-4)
Document the generator dry-run and apply outcomes for task 15: 42 skills
generated (22 API + 20 CLI), 18 omniroute-* orphans pruned to archive,
idempotency verified, and 10 custom-block skill files confirmed preserved.
2026-05-28 02:04:44 -03:00
diegosouzapw
d9f4035d2c refactor(agent-skills): remove deprecated AGENT_SKILLS dead code (GAP-3)
AgentSkillsPageClient (F7) was rewritten and no longer imports AGENT_SKILLS.
Grep confirms zero remaining imports of the symbol in src/, open-sse/, tests/,
or electron/. Remove the backward-compatible shim and TODO(F7) comment.
2026-05-28 02:00:44 -03:00
diegosouzapw
9f88f39f6a docs(skills): rewrite README.md with 42 new skill IDs (GAP-2)
Replace the stale 18-entry omniroute-* index (with broken links to pruned
skill directories) with a complete table of all 42 current skills (22 API +
20 CLI). Adds entry-point pointers, raw URL pattern, agent-discovery section
(MCP tool + A2A skill), and cross-reference to docs/frameworks/AGENT-SKILLS.md.
2026-05-28 01:57:38 -03:00
diegosouzapw
4797c08018 fix(agent-skills): use res.text() for text/markdown endpoint (GAP-1 critical bug)
The /api/agent-skills/[id]/raw endpoint returns text/markdown (plain string),
not a JSON envelope. The client was incorrectly calling res.json() and
unpacking a non-existent `body` field, causing all preview panes to show
empty content. Switch to res.text() and update the test mock to match.
2026-05-28 01:53:59 -03:00
diegosouzapw
5c79c1a32d merge(F10): final audit + docs + integration tests + openapi 2026-05-28 01:35:30 -03:00
diegosouzapw
b54de3278a docs(openapi): add /api/agent-skills/* paths and AgentSkill + SkillCoverage schemas
Add tag: Agent Skills (with description of the 42-entry catalog).
Add 5 paths:
- GET /api/agent-skills (list with category/area filters)
- GET /api/agent-skills/{id} (single skill metadata)
- GET /api/agent-skills/{id}/raw (SKILL.md as text/markdown)
- GET /api/agent-skills/coverage (coverage stats)
- POST /api/agent-skills/generate (requires management auth, dryRun/prune/onlyIds body)

Add 2 schemas:
- AgentSkill: id, name, description, category, area, endpoints, cliCommands,
  icon, isEntry, isNew, rawUrl, githubUrl
- SkillCoverage: api{have,total:22}, cli{have,total:20}, totalSkills, generatedAt

Add ErrorResponse schema (reusable for agent-skills error responses).
2026-05-28 01:10:44 -03:00
diegosouzapw
a430039ef0 docs(architecture): add src/lib/agentSkills entry in CODEBASE_DOCUMENTATION
Add agentSkills/ row to the src/lib/ module table:
- catalog.ts: getCatalog/getSkillById/filterCatalog/computeCoverage
- generator.ts: generateAgentSkills (writes skills/{id}/SKILL.md)
- openapiParser.ts, cliRegistryParser.ts, schemas.ts, types.ts
- Consumed by REST routes, MCP tools, and A2A list-capabilities skill

Update a2a/ row: 5 → 6 skills (add list-capabilities).
2026-05-28 01:10:31 -03:00
diegosouzapw
914583d580 docs(a2a): add list-capabilities skill entry
Update Available Skills table from 5 to 6 skills.
Add list-capabilities skill with full columns: ID, description, tags, examples.
Add detail subsection explaining the markdown table artifact format,
rawUrl column, metadata.totalSkills, and link to AGENT-SKILLS.md.
2026-05-28 01:10:18 -03:00
diegosouzapw
2ee80c6702 docs(mcp): add omniroute_agent_skills_* tools + read:catalog scope
Add "Agent Skill Catalog Tools (3)" section after Skill Tools (4):
- omniroute_agent_skills_list (read:catalog)
- omniroute_agent_skills_get (read:catalog)
- omniroute_agent_skills_coverage (read:catalog)

Add read:catalog row to Authentication & Scopes table.
Update count in section intro (40 tools total, including 3 new catalog tools).
2026-05-28 01:10:03 -03:00
diegosouzapw
712b00346c docs(agent-skills): add AGENT-SKILLS.md + cross-link in SKILLS.md
Add docs/frameworks/AGENT-SKILLS.md (~190 lines):
- Overview of the 42-entry dynamic catalog (22 API + 20 CLI)
- Architecture map: catalog.ts, generator.ts, openapiParser.ts,
  cliRegistryParser.ts, schemas.ts, types.ts
- SKILL.md format with custom-block documentation
- REST API discovery reference table
- MCP tools table (omniroute_agent_skills_list/get/coverage)
- A2A list-capabilities invocation example
- Full 42-ID catalog tables (API + CLI)
- External agent consumption guide (REST/MCP/A2A/GitHub raw)
- Generator usage (dryRun/prune/onlyIds)
- Coverage API reference

Update docs/frameworks/SKILLS.md:
- Add "## Agent Skills vs Omni Skills" section at top with comparison table
- Link to AGENT-SKILLS.md
2026-05-28 01:09:47 -03:00
diegosouzapw
e7a1bf1c24 test(agent-skills): integration tests for discovery + content + MCP + A2A
Add tests/integration/agent-skills-discovery.test.ts (13 tests):
- Verifies every API_SKILL_IDS + CLI_SKILL_IDS has skills/<id>/SKILL.md on disk
- Validates frontmatter name + description present in each SKILL.md
- Validates body >= 100 chars per SKILL.md
- Verifies MCP omniroute_agent_skills_list handler returns 42 entries
- Verifies A2A list-capabilities returns 1 artifact containing all 42 IDs

Add tests/integration/agent-skills-content.test.ts (16 tests):
- Confirms all 42 catalog IDs have skills/{id}/ directory + SKILL.md
- Confirms 0 omniroute-* directories remain (post-prune)
- Confirms exactly 10 IDs have skill:custom-start/end blocks
  (omni-mcp, omni-compression, cli-providers, cli-eval, omni-agents-a2a,
   omni-combos-routing, omni-auth, omni-resilience, omni-inference, cli-serve)
- Confirms no unclosed custom blocks
2026-05-28 01:09:32 -03:00
diegosouzapw
8b430bfcc9 merge(F9b): apply generator + prune (42 SKILL.md, 18 omniroute-* pruned) 2026-05-27 23:51:38 -03:00
diegosouzapw
f4571094c2 feat(skills): generate 42 SKILL.md (22 API + 20 CLI) + prune 18 omniroute-* orphans 2026-05-27 23:50:13 -03:00
diegosouzapw
591d99ca36 merge(F9a): preserve curated content from 18 old SKILL.md via markers + archive 2026-05-27 23:41:08 -03:00
diegosouzapw
2e1862192e feat(skills): preserve curated content from 18 old SKILL.md via custom-start/end markers + archive
Migrates 1560 lines of curated content from 18 omniroute-* skill files
into 10 new destination SKILL.md placeholders using the generator-safe
<!-- skill:custom-start --> ... <!-- skill:custom-end --> markers so
F9b --apply preserves them instead of overwriting.

Mappings:
- omniroute → omni-auth (direct)
- omniroute-chat + image + tts + stt + embeddings + web-search + web-fetch → omni-inference (aggregated, 7 sections)
- omniroute-mcp → omni-mcp (direct)
- omniroute-a2a → omni-agents-a2a (direct)
- omniroute-routing → omni-combos-routing (direct)
- omniroute-compression → omni-compression (direct)
- omniroute-monitoring → omni-resilience (direct)
- omniroute-cli + omniroute-cli-admin → cli-serve (aggregated, 2 sections)
- omniroute-cli-providers → cli-providers (direct)
- omniroute-cli-eval → cli-eval (direct)
- omniroute-cli-cloud → archived only (no direct equivalent)

Archive: all 18 originals copied to
_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/
2026-05-27 23:39:44 -03:00
diegosouzapw
2b47a81d54 merge(F7): integrate W3 frente F7 into base 2026-05-27 22:17:26 -03:00
diegosouzapw
0e357a9cc7 merge(F6): integrate W3 frente F6 into base 2026-05-27 22:17:24 -03:00
diegosouzapw
95312401b6 merge(F5): integrate W3 frente F5 into base 2026-05-27 22:17:23 -03:00
diegosouzapw
1d44f5d35d merge(F4): integrate W3 frente F4 into base 2026-05-27 22:17:21 -03:00
diegosouzapw
8ed2c02808 merge(F3): integrate W3 frente F3 into base 2026-05-27 22:17:20 -03:00
diegosouzapw
14d4c7cbcd test(agent-skills): unit tests for AgentSkillsPageClient + filters + selection 2026-05-27 22:10:19 -03:00
diegosouzapw
d27b86568b feat(agent-skills): wire AgentSkillsPageClient with split grid + preview + actions 2026-05-27 22:10:14 -03:00
diegosouzapw
1aaf43d89e feat(agent-skills): add SkillCard, SkillPreviewPane, CoverageBar, McpA2aLinksBar 2026-05-27 22:10:08 -03:00
diegosouzapw
dff8524d3d refactor(agent-skills): rewrite dashboard page as server wrapper + client 2026-05-27 22:10:01 -03:00
diegosouzapw
4a97b419ae test(mcp): unit tests for agentSkillTools 2026-05-27 21:58:22 -03:00
diegosouzapw
a7a093d066 feat(mcp): register read:catalog scope + add tools to MCP_TOOLS array 2026-05-27 21:58:18 -03:00
diegosouzapw
d96e74bc15 feat(mcp): add agentSkillTools (list/get/coverage) for agent-skills discovery 2026-05-27 21:58:14 -03:00
diegosouzapw
d1d0ddda0f test(agent-skills): unit tests for generator + scripts smoke
20 tests covering:
- dry-run produces report without writing (filesystem unchanged)
- apply writes SKILL.md for API and CLI skills with correct sections
- apply writes all 42 skills when no filter
- idempotency: second run reports 0 generated, all unchanged
- prune dry-run detects orphans without deleting
- prune apply deletes orphan dirs, preserves catalog dirs
- marker preservation: custom block survives regeneration
- buildSkillMarkdown: valid shape, no escape errors, correct sections per category
- onlyIds filter limits generation
- report shape validation

Also adds gray-matter as dependency (per plan prerequisite check).
2026-05-27 21:55:01 -03:00
diegosouzapw
ca41880bb3 feat(agent-skills): add CLI wrapper scripts/skills/generate-agent-skills.mjs
Implements the CLI wrapper for the generator:
- --apply flag: enables write mode (default: dry-run)
- --prune flag: enables orphan detection/deletion
- --only=<id1,id2>: filters to specific skill IDs
- --json flag: structured JSON output
- Exit codes: 0 success, 1 error, 2 dry-run detected pending changes (CI)
- Human-readable table output with Generated/Unchanged/Pruned/Orphans/Errors summary
- Dynamic import via tsx/esm runtime; falls back to module.register() if needed
2026-05-27 21:54:50 -03:00
diegosouzapw
f86e905efb feat(agent-skills): add generator with idempotent skill md generation + prune
Implements src/lib/agentSkills/generator.ts (§3.4 contract):
- generateAgentSkills(opts): idempotent, dryRun:true default, prune:false default
- buildSkillMarkdown(skillId, sources): generates frontmatter + API/CLI body
- API body: Visão geral + Autenticação + Endpoints (with curl) + Payloads
- CLI body: Visão geral + Instalação rápida + Subcomandos (with flags + examples)
- Prune: detects orphans in skills/{id}/ not in catalog; deletes in apply mode
- Marker preservation: <!-- skill:custom-start --> ... <!-- skill:custom-end -->
- Generated comment: per D24 spec
2026-05-27 21:54:42 -03:00
diegosouzapw
b7efba4727 test(agent-skills): unit tests for REST routes with sanitization assertions
16 tests covering all 5 endpoints:
- GET /agent-skills: 42 total, category/area filters, invalid category → 400
- GET /agent-skills/[id]: found api+cli skills, unknown → 404
- GET /agent-skills/[id]/raw: unknown → 404, valid → 200/502/500 (network-tolerant)
- GET /agent-skills/coverage: SkillCoverage shape (api.total=22, cli.total=20)
- POST /generate: no auth → 401/403, invalid body → 400, no generator → 503

Hard Rule #12 verified explicitly in every error case: error.message must
not match /\bat \/|\bat file:\/\// (no stack trace exposure). Final test
collects all error paths and asserts sanitization in a single sweep.
2026-05-27 21:49:32 -03:00
diegosouzapw
ccda3cf0f0 feat(agent-skills): add POST /generate with management auth + dryRun default
Implements POST /api/agent-skills/generate (F4):

- Requires management auth via requireManagementAuth (same pattern as
  usage/combo-health-dashboard and other management routes)
- Validates body via GenerateBodySchema (dryRun defaults true, prune
  defaults false — safe preview mode)
- Dynamic import of @/lib/agentSkills/generator so the route coexists
  before F3 (generator.ts) is merged; returns 503 if module unavailable
- Hard Rule #12: all error paths use buildErrorBody (400/401/403/503/500)
- Hard Rule #7: Zod validates both JSON parse and schema shape
2026-05-27 21:49:21 -03:00
diegosouzapw
d1160120c0 feat(agent-skills): add REST GET endpoints (list, get, raw, coverage)
Implements four read-only endpoints for the Agent Skills REST API (F4):

- GET /api/agent-skills — catalog with ?category= and ?area= filters,
  returns { skills, count, coverage }
- GET /api/agent-skills/[id] — single skill by canonical ID (404 if absent)
- GET /api/agent-skills/[id]/raw — SKILL.md as text/markdown with
  Cache-Control: public, max-age=3600; 502 on GitHub fallback failure
- GET /api/agent-skills/coverage — SkillCoverage (filesystem vs catalog)

All routes: export const dynamic = "force-dynamic" (filesystem reads),
error responses via buildErrorBody (Hard Rule #12), Zod input validation
(Hard Rule #7). coverage/route.ts force-added to git because .gitignore
matches the directory name "coverage/" — this is an API route, not a
report directory.
2026-05-27 21:49:12 -03:00
diegosouzapw
754806e0f8 test(a2a): unit tests for listCapabilities + Agent Card route
Adds 6 tests for executeListCapabilities (§3.7 shape, 42-skill IDs in markdown,
coverage bounds, ISO datetime, handler registration) and 5 tests for the Agent
Card route (6 skills total, list-capabilities presence + tags + examples, all
original 5 skills preserved). All 11 pass.
2026-05-27 21:49:03 -03:00
diegosouzapw
c91decf543 feat(a2a): register list-capabilities in A2A_SKILL_HANDLERS + Agent Card
Adds "list-capabilities" entry to A2A_SKILL_HANDLERS in taskExecution.ts
(dynamic import pattern, consistent with the 5 existing skills) and adds
the 6th skill entry to /.well-known/agent.json with tags [discovery, capabilities]
and example questions for agent discovery.
2026-05-27 21:48:57 -03:00
diegosouzapw
6bad368dcb feat(a2a): add list-capabilities skill with markdown table of 42 skills
Implements executeListCapabilities() which calls getCatalog() + computeCoverage()
from the agentSkills catalog (F1/F2) and returns a markdown table covering all
42 skills (22 API + 20 CLI) with ID, name, category, area, endpoints/commands,
and raw SKILL.md URL, matching the §3.7 result contract.
2026-05-27 21:48:50 -03:00
diegosouzapw
44b248314b merge(F8): move skills→omni-skills + split into 8 components + inspector 2026-05-27 21:26:38 -03:00
diegosouzapw
f730161c74 merge(F2): catalog source 42 entries + parsers OpenAPI/CLI 2026-05-27 21:26:37 -03:00
diegosouzapw
b1a127a732 test(omni-skills): unit tests for OmniSkillsPageClient and components
46 structural/source-pattern tests covering: file structure, page.tsx server
component contract, PageClient 4-tab wiring, OmniSkillCard accessibility,
SkillInspectorPane 4 sub-tabs + API fetches, OmniSkillsList split layout,
OmniExecutionsTab columns, OmniSandboxTab config values,
OmniMarketplaceTab dual-provider logic, and E2E path update.
2026-05-27 21:24:15 -03:00
diegosouzapw
33c9b2b96c test(omni-skills): update e2e path /dashboard/skills → /dashboard/omni-skills 2026-05-27 21:24:10 -03:00
diegosouzapw
99ccc35b93 refactor(omni-skills): split monolithic page into PageClient + 6 components
Extract the 870-line page.tsx into OmniSkillsPageClient (state + orchestration)
and 6 focused components: OmniSkillCard, OmniSkillsList, SkillInspectorPane,
OmniExecutionsTab, OmniSandboxTab, OmniMarketplaceTab.

- SkillsConceptCard variant="omni" rendered at top of page
- 4 tabs (skills/executions/sandbox/marketplace) preserved
- Skills tab uses 12-col split grid with SkillInspectorPane (4 sub-tabs)
- SkillInspectorPane: schema/handler/executions/sandbox sub-tabs, mode buttons,
  uninstall, per-skill execution fetch
2026-05-27 21:24:06 -03:00
diegosouzapw
dfd2182c41 refactor(omni-skills): move dashboard/skills → dashboard/omni-skills
Rename the page route from /dashboard/skills to /dashboard/omni-skills
to align with the task-15 F8 redesign plan. The server component wrapper
is reduced to a 5-line delegator, removing the prior 870-line monolith.
2026-05-27 21:23:55 -03:00
diegosouzapw
15838348c3 test(agent-skills): unit tests for catalog + openapi + cli parsers
agentSkills-catalog.test.ts (30 tests):
- getCatalog(): 42 total, 22 api, 20 cli
- API_SKILL_IDS/CLI_SKILL_IDS length assertions
- ID regex format, uniqueness, required fields
- getSkillById happy path + null for unknown/empty
- filterCatalog by category, area, combined, empty
- refreshCatalog() invalidates cache (new array reference)
- computeCoverage() shape validation
- rawUrl/githubUrl URL format assertions

agentSkills-openapiParser.test.ts (9 tests):
- Fixture YAML: paths Map, area groupings (providers, api-keys, inference)
- OpenapiPath field validation
- Missing file throws
- Empty paths YAML returns empty Maps
- Real openapi.yaml: providers area ≥5 endpoints (integration)

agentSkills-cliRegistryParser.test.ts (10 tests):
- Fixture .mjs: commands Map, families Map, ≥5 provider subcommands
- Description extraction, isSubcommand flag, flags extraction
- Skips unrecognised files, throws on missing dir
- Real providers.mjs: ≥5 commands (integration)
2026-05-27 20:32:08 -03:00
diegosouzapw
0a95372746 feat(agent-skills): add openapiParser and cliRegistryParser
openapiParser.ts:
- parseOpenapi(): reads docs/reference/openapi.yaml via js-yaml (already a dep)
  and returns { paths: Map<METHOD+path, OpenapiPath>, areas: Map<SkillArea, ops[]> }
- PATH_AREA_MAP maps 30+ path prefixes to SkillArea values
- getEndpointsForArea(area): convenience helper returning 'METHOD /path' strings

cliRegistryParser.ts:
- parseCliRegistry(): reads all bin/cli/commands/*.mjs via fs.readdirSync
  and regex-parses .command(), .description(), .option() calls
- FILE_FAMILY_MAP maps 40+ file basenames to CLI SkillArea families
- getCommandsForFamily(family): convenience helper for catalog consumers
- Does NOT import Commander.js modules to avoid side-effects (D15)
2026-05-27 20:31:55 -03:00
diegosouzapw
a0cc22be73 feat(agent-skills): add catalog.ts with getCatalog/filter/coverage/fetch helpers
Implements the catalog.ts public API defined in §3.3 of the master plan:
- getCatalog(): AgentSkill[] — returns 42 entries, lazy-cached in module scope
- getSkillById(id): AgentSkill | null — lookup by canonical ID
- filterCatalog(opts): AgentSkill[] — filter by category and/or area
- computeCoverage(): SkillCoverage — reads skills/ dir and counts SKILL.md present
- refreshCatalog(): void — invalidates cache (used by tests + generator)
- fetchSkillMarkdown(id): Promise<SkillMarkdown> — reads local fs first,
  falls back to GitHub raw fetch with 1h Next.js cache (for F4 /raw route)

API_SKILL_IDS and CLI_SKILL_IDS exported as readonly string arrays (D28 order).
Single source of truth for all consumers (REST routes, MCP, A2A).
2026-05-27 20:31:44 -03:00
diegosouzapw
78de1d2455 feat(agent-skills): expand catalog source to 42 entries (22 API + 20 CLI)
Replace 18-entry hardcoded AGENT_SKILLS array with 42-entry CURATED_SKILLS
covering all areas listed in D28 of the master plan. Each curated entry
has id, name, description, category, area, icon, and optional flags.

Adds backward-compatible AGENT_SKILLS alias (deprecated) that maps curated
entries to the old AgentSkill shape with empty endpoints/cliCommands arrays
so the existing /dashboard/agent-skills page continues to work until F7
rewrites it.

Imports AgentSkill, SkillArea, SkillCategory types from src/lib/agentSkills/types.ts
(F1 single source of truth) instead of redeclaring locally.
2026-05-27 20:31:31 -03:00
diegosouzapw
bf764dc529 test(agent-skills): unit tests for schemas + SkillsConceptCard 2026-05-27 19:23:36 -03:00
diegosouzapw
612cf63de7 feat(agent-skills): redirect /dashboard/skills -> /dashboard/omni-skills + sidebar reorder 2026-05-27 19:23:32 -03:00