feat: consolidação KonectySkills — três skills, README bilíngue, LICENSE AGPL-3.0 e publicação#2
Open
silveirado wants to merge 44 commits into
Open
feat: consolidação KonectySkills — três skills, README bilíngue, LICENSE AGPL-3.0 e publicação#2silveirado wants to merge 44 commits into
silveirado wants to merge 44 commits into
Conversation
7 docs covering stack, architecture, conventions, structure, testing, integrations, and concerns — generated via tlc-spec-driven brownfield mapping. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Decision log from grilling session. Two self-sufficient skills (konecty-data + konecty-meta) replace 18 separate skills, with shared auth/modules enforced via manifest-based gates. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…CONS-01, CONS-03) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…very.md (CONS-01, CONS-03) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…very.md (CONS-03) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ONS-05) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
All 18 original skills covered — zero missing entries. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…nd konecty-meta - 17 tests for konecty-data (modules, find, CRUD lifecycle with write ops) - 15 tests for konecty-meta read/doctor - 17 tests for remaining meta ops (document, list, view, access, hook, namespace, pivot, sync, remove) including write-op lifecycle tests (add-field, upsert-list, upsert/delete-hook) - run_coverage.sh generates text/HTML/XML reports (39% total, up to 79% per script) - fix: meta_namespace._get_namespace() uses dynamic discovery to handle instances where _id != compound pattern (avoids 404 → SystemExit) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Consolidation complete: replace 18 individual skill directories with the two consolidated skills (konecty-data, konecty-meta). Update AGENTS.md API surface table and skills map to reflect the new structure. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
All 16 tasks in the konecty-skills-consolidation feature are now done. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…r marketing skills - AGENTS.md: add Read first + Commands sections, completion-gate workflow (codebase-intelligence/security), available-skills table, and the shared-files byte-identical invariant - Makefile: expand from single setup target to self-documenting set (help, lint, shared-check, validate, test, test-cov, audit, check, clean) - .agents/skills: vendor copywriting, content-strategy, marketing-ideas, marketing-psychology from coreyhaines31/marketingskills - skills-lock.json: lock the 4 marketing skills + previously-untracked codebase-intelligence and codebase-security (10 total) - .gitignore: ignore .coverage and .serena tooling artifacts - docs/changelog: new entry + README row Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…mplexity The codebase-intelligence gate failed on two pre-existing CC>=25 functions and reported a long tail of findings inside vendored .agents/ skills this repo does not own. After this change: dead_code=0, critical_cc=0, intelligence FAIL->WARN. - ruff.toml: extend-exclude=[".agents"] (dead-code layer scans the whole repo) - both audit.sh (vendored): drop dot-dir paths from --changed-since targets, mirroring the full-scan branch — keeps .agents out of bandit/SAST scope - upload.py: cmd_upload CC 33->6 (extract validate/extract-metadata/print) - meta_remove.py: cmd_apply CC 31->11 (extract guard/process-item/warn-state), drop unused `skipped` counter and unused `name` unpack (dead code) - ruff check --fix skills/: clear unused import + redundant f-string prefixes - .gitignore: ignore .codebase-audit/ and .security-audit/ output dirs Remaining WARN is by design: shared-files duplication (auth.py/modules.py kept byte-identical across skills) and B310 SAST false-positives on urllib.urlopen. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…nstall Implements T8 (install + idempotent entry-file merge) and T9 (update preserving local edits, uninstall with conflict guards). Manifest records per-file sha256 under composite <engine>:<skill> keys. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…or/update/uninstall) Implements T10–T13. Adds a KONECTY_HOME path seam for test isolation, full install flow (banner→detect→fetch→copy→merge→manifest→credentials), and the configure/status/doctor/update/uninstall commands. 128 unit tests green. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements T14. README quick-install section, AGENTS.md structure+commands, make installer-test (folded into make check), and a changelog entry. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements T15. Integration validation (install/update/uninstall against real skill content) ALL PASS. Documents the merge-to-main blocker and the uvx local-path caching gotcha. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…B105) Remove dead imports/vars/_stub after wiring; add https/scheme guards before urlopen (fetcher, doctor probe); annotate intentional subprocess + None-dict with nosec. Installer is audit-clean (0 security, 0 dead-code findings). 135 tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
python3 -m konecty_skills.globe [height] renders a shaded ASCII sphere homage to the brand logo (angular palette + sqrt(1-r^2) shading; plain density fallback for non-TTY/NO_COLOR). 5 unit tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Globe redrawn as a spherical Voronoi (Fibonacci seeds, brand-palette cells, white channels) to match the logo's puzzle-piece look. banner.full()/print_full() stack a compact globe above the KONECTY wordmark; install now shows it. 143 tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
print_full (globe + wordmark) is the sole banner entry; render() remains for composition. Repointed NO_COLOR/TTY tests and install-flow patches. 143 tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Spin up a disposable Konecty stack (mongo rs0 + init + rabbitmq + konecty/konecty:3.8.10 on alt ports) and drive every skill subcommand via an in-process pseudo-agent, reaching 93% line coverage (gate --fail-under=90). `make e2e` is fully self-contained: it purges any prior stack, boots a clean one, waits for health, runs the suite under the coverage gate, and ALWAYS tears down with `down -v` afterwards (trap on EXIT/INT/TERM) — so a failed or interrupted run never leaves containers/volumes behind to contaminate the next. Hybrid architecture (see .specs/project/STATE.md D6-D10): - konecty-data: tested LIVE against 3.8.10 for the subset the public image supports (auth, modules, find, create, update-by-ids). - konecty-meta + drifted data paths (query/json now needs relations>=1; /api/admin/meta/* not in a published image — Konecty PR #299): covered by a faithful in-memory MockKonecty. Live swap deferred to a PR-299 image. Adds: - e2e/docker-compose.yml + bootstrap scripts (admin token from logs, liveness) - tests/e2e/: pseudo-agent, reporter, deterministic PT/EN intent router, MockKonecty, and suites (live, mock data+meta, security, inference, closers) - tests/conftest.py, pytest.ini, .coveragerc - make e2e* targets (e2e, e2e-up/down/reset/token/run/cov/sec/infer) - SDD specs (.specs/features/e2e-harness/, .specs/project/STATE.md), changelog - untrack stale generated coverage artifacts; gitignore coverage reports 472 passed, 1 documented xfail. make check + make audit clean (no fail verdict). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Third Konecty skill, first advisory one (no scripts, no live calls): teaches developer-agents to write integration code. SDK-first (Python konecty_sdk_python 2.0.3, Node/TS @konecty/sdk 1.0.0) with the full REST API documented as a first-class agnostic track for languages without an SDK. - SKILL.md: lean router (bilingual description, trigger table, SDK->REST cascade) - 8 references: getting-started, auth-for-code, python-sdk, typescript-sdk, rest-api, filters, recipes, hooks - hooks.md: runtime contract of the 3 data hooks (write-the-logic cut) - Out of the shared-files invariant (advisory; distinct service-account auth doc) - Specs (spec/design/tasks), ADR-0005 (patterns-not-content), changelog - AGENTS.md: pt-BR reply preference; .gitignore: protect *.local.md source maps Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace a real client name/domains in konecty-meta reference examples and one changelog with generic placeholders (acme, *.example.com). The repo is public; examples must not carry client-identifying content. Working-tree cleanup only — history purge is a separate force-push step. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…rewritten) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
silveirado
force-pushed
the
feat/konecty-skills-consolidation
branch
from
b99ef3b to
a7c1587
Compare
- README.md (pt-BR) e README.en.md (EN): logo adaptivo claro/escuro, badges reais (Snyk ao vivo, cobertura E2E, AGPL), seções completas de instalação (uvx, skills.sh, OpenClaw, Hermes, manual), as três skills com tabelas de trigger phrases, credenciais, segurança com datas de verificação, testes E2E, estrutura e publicação - LICENSE: AGPL-3.0 (alinhado com o repo da plataforma Konecty) - Makefile: variáveis DEV + ALL_SKILLS, targets publish-gh / publish-clawhub / publish-hermes / publish com VERSION e CHANGELOG configuráveis; validate agora cobre konecty-dev também - docs/publishing.md: guia passo a passo para GitHub, skills.sh, OpenClaw, Hermes, Anthropic/skills, Snyk badge ao vivo e auditorias manuais (Gen Agent Trust Hub, Socket, Snyk agent scan) - docs/README.md + development.md: links para publishing.md Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
silveirado
changed the title
Skills externas instaladas via CLI ficam apenas no ambiente local de cada desenvolvedor — não faz sentido versionar junto ao repo. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
silveirado
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
O que esse PR inclui
konecty-dev (skill advisory — já implementada)
Terceira skill: ensina agentes a escrever código de integração com Konecty (SDK Python, SDK TypeScript/Node, API REST). Puramente advisory — nenhum script ao vivo. Lean
SKILL.md+ 8 referências (~2.5k linhas).Documentação
uvx+ skills.sh + OpenClaw + Hermes + manual, as três skills com tabelas de trigger phrases, credenciais, segurança, testes E2E, estrutura e publicaçãoLICENSE
AGPL-3.0 — alinhado com o repositório da plataforma Konecty.
Makefile
DEV+ALL_SKILLS(cobre konecty-dev nas publicações)publish-gh/publish-clawhub/publish-hermes/publishcomVERSIONeCHANGELOGconfiguráveisvalidateagora cobre as três skillsSegurança (badges)
snyk.io/test/github/konecty/skills) — ativa após importar o repo no snyk.ioTest plan
make check— lint + shared-files + installer tests passandomake lint— py_compile OKgit push— branch atualizada no remotomake publish-gh VERSION=x.y.zapós merge em main🤖 Generated with Claude Code