convez.ai
Skill
hoanganhduc

ai-agents-skills

hoanganhduc/ai-agents-skills

Ver no GitHub
InicialPluginScore · de 0 a 100

Shared skills and settings installer for Codex, Claude, and DeepSeek

Instalação

$npx convez-ai hoanganhduc/ai-agents-skills

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
2
Forks
1
Acompanhando
1
Saúde
42%
Último commit
há 4 dias
Versão
—
Issues abertas
0
Contribuidores
3

README

A tradução para o português deste README ainda está na fila.

AI Agents Skills

Python Platforms Agents Docs Status License

⚠️ Generated by an AI coding system — experimental. This repository's code, configuration, and documentation were generated by the same multi-agent AI coding system it installs skills for. It is experimental and may not work as expected outside the author's setup. For an overview of the whole system and how to try it, see the introductory blog post.

Shared, manifest-driven skills and settings for Codex, Claude, DeepSeek, GitHub Copilot, OpenCode, Antigravity CLI, and restricted OpenClaw fake-root targets.

System Summary

This is an experimental, personal-use configuration for research workflows, especially combinatorics and graph theory work. It is not a polished general product, and it may not behave as desired on other machines, other agent versions, or research tasks outside the assumptions documented here.

This repo turns a multi-agent research setup into one maintainable skill source. Codex, Claude, DeepSeek, GitHub Copilot, OpenCode, and Antigravity CLI can each load local skills. OpenClaw participates as a default fake-root-only target for normal installer flows and has a separate reviewed v2 skill-file path for real-system target writes. This repository keeps the shared research workflows, profiles, delegation settings, dependency metadata, and installer logic in one place.

The research stack is organized as:

  • agent frontends and targets: Codex, Claude, DeepSeek, GitHub Copilot, OpenCode, Antigravity CLI, and restricted OpenClaw
  • shared skill source: manifest/, canonical/skills/, and targets/
  • external capabilities: Python, TeX, optional SageMath, local library tools, document parsers, public databases, and retrieval helpers

For example, a literature-review request can route through research-briefing, deep-research-workflow, paper-lookup, and research-verification-gate; a paper-review request can check zotero first, fall back to calibre for books, parse files with docling, and then run paper-review.

Nontrivial research workflows are expected to keep evidence control visible: scope briefs define claim boundaries and goal/backward success checks, deep-research runs preserve source handoffs and guard outputs, delegated agents return bounded result packets, and final reports pass review and verification gates. The workflow tracks concrete issues and evidence gaps instead of hiding quality behind a single aggregate score.

Reusable failures, corrections, and missing capabilities route through self-improving-agent. That skill logs local .learnings/ entries and, when a lesson affects skills or settings, proposes a canonical repo integration plan that names affected install targets, OS/substrates, docs, manifests, runtime helpers, tests, and blocked coverage before any files are changed.

See docs/workflow-overview.md for the full sanitized system description and workflow examples.

Multi-agent work is documented separately in docs/multi-agent-examples.md. That page explains how the orchestrator selects templates, spawns bounded role agents, waits for round outputs, runs verification, and merges the result. It also summarizes the available templates:

  • Lakatos Proof and Refutation: proof stress-testing.
  • Polya Multi-Strategy Problem Solving: open problem exploration.
  • Knuth Structured Manuscript Review: mathematical draft review.
  • Structured Research Team: high-stakes claim and proof review.
  • Graph Reconfiguration Specialist: gadgets, reductions, and PSPACE/NP-hardness checks.
  • Lean Formalization Team: Lean skeleton and proof-blocker analysis.
  • Prose / OpenProse-style workflow: reproducible decomposition and synthesis.

This repo is a generator and installer, not a copied dotfiles folder. It uses canonical skill names, generates per-agent adapters, supports partial installs, detects legacy/self-contained installs, and verifies only installed managed skills. Reusable skill bodies live under canonical/skills; default installs record the agent policy and fallback behavior used to choose symlink, reference, or copy mode. Copy mode remains available when an agent or filesystem must have regular files inside the settings directory.

Platform support is Linux and Windows first, with core installer flows also covered on macOS in CI. macOS users should expect POSIX-style behavior but lighter platform-specific guidance than Linux and Windows.

Documentation

The GitHub Pages site is built from docs/source and deployed by .github/workflows/docs.yml.

Most checked-in docs are generated. Edit installer/ai_agents_skills/docs.py and the manifests for generated pages, then run make docs; CI checks that generated docs are current. Generated docs are README.md, each page emitted by generated_doc_texts() under docs/, and the mirrored copies under docs/source/. docs/source/index.md, docs/source/overview.md, and docs/source/submission-venue-selector-plan.md are maintained manually; docs/submission-venue-selector-plan.md is the top-level manual copy of the same plan.

Acknowledgements

This repository was implemented and maintained with help from ChatGPT Codex.

License

This project is licensed under GPL-3.0-or-later. See LICENSE.

Quick Start

Clone the repo and run commands from the repository root:

git clone https://github.com/hoanganhduc/ai-agents-skills.git
cd ai-agents-skills

Requires Python 3.10 or newer. Linux and macOS examples use make and the POSIX bootstrap script. Windows examples use make.bat, which requires pwsh or powershell.exe. The installer only plans targets for existing agent homes; absent homes are reported and skipped.

Linux/macOS:

make doctor
make precheck ARGS="--profile research-core"
make audit-system ARGS="--profile research-core"
make list-skills
make list-artifacts
make plan ARGS="--profile research-core"
make plan ARGS="--no-skills --artifact-profile workflow-templates"
make install ARGS="--profile research-core --dry-run"
make lifecycle-test ARGS="--matrix default --platform-shape all"
make fake-root-lifecycle ARGS="--profile research-core --platform-shape linux"

For a macOS-shaped fake-root check, use --platform-shape macos in the final command.

Windows:

make.bat doctor
make.bat precheck --profile research-core
make.bat audit-system --profile research-core
make.bat list-skills
make.bat list-artifacts
make.bat plan --profile research-core
make.bat plan --no-skills --artifact-profile workflow-templates
make.bat install --profile research-core --dry-run
make.bat lifecycle-test --matrix default --platform-shape windows
make.bat fake-root-lifecycle --profile research-core --platform-shape windows

For a shorter first pass, run doctor, precheck, plan, and a dry-run install before any lifecycle matrix. lifecycle-test and fake-root-lifecycle are verification tools; they use fake roots and should not be confused with a real install.

Applied installs, uninstalls, and rollbacks are interactive: before any --apply writes files, the installer explains the install, uninstall, and rollback process and requires the user to type the displayed confirmation phrase. Real-system writes also require explicit --apply --real-system. Tests and examples use fake roots. Existing unmanaged files are skipped by default; use --adopt, --backup-replace, or --migrate only after reviewing plan output.

After reviewing plan output, a real install uses both write gates:

make install ARGS="--profile research-core --apply --real-system"

Skills install in --install-mode auto by default so the repo remains the single maintained source without hiding agent-loader differences. plan --json shows the effective mode, agent policy evidence, apply-time symlink fallback, and reason for each target. Use --install-mode symlink to force symlinks for every agent, --install-mode reference to force adapters for every agent, or --install-mode copy only when files must be materialized inside the agent settings directory.

Optional workflow artifacts are not installed by default. Use --artifact-profile workflow-templates, --artifact-profile review-personas, --artifact-profile workflow-instructions, or --artifact-profile research-entrypoints explicitly. Use --with-deps when dependency-bound artifacts should also install their backing skills.

Command Surfaces

  • make <target> ARGS="..." is the normal Linux/macOS wrapper.
  • make.bat <command> ... is the normal native Windows wrapper.
  • ./installer/bootstrap.sh <command> ... and python3 -m installer.ai_agents_skills <command> ... are direct POSIX entrypoints for installer CLI commands when debugging wrapper behavior.
  • Installer CLI commands include doctor, precheck, audit-system, plan, install, verify, smoke, rollback, uninstall, runtime-smoke, lifecycle-test, list-skills, list-artifacts, describe, and describe-artifact.
  • Makefile-only maintainer targets include docs, docs-site, docs-check, static-check, sanitize-check, test, and release-check; run them through make or make.bat, not as installer CLI commands.

Runtime-Backed Skills

Runtime-backed skills install shared helper files from canonical/runtime into a root-scoped runtime directory instead of copying executable helpers inside every agent's skill directory. The runtime manifest declares platform filters, newline policy, executable modes, and the exact source-to-target mapping. Runtime inventory intentionally rejects live config, databases, caches, downloaded documents, bytecode, archives, symlinks, sensitive material, and persistent execution markers.

Docling is the main document/OCR runtime-backed skill. Its managed wrapper is local-only by default: sources must be local files and remote service fields are rejected from config. Use scan-heavy when you want stronger local OCR for image-backed papers:

bash ~/.codex/runtime/run_skill.sh skills/docling/run_docling.sh doctor
bash ~/.codex/runtime/run_skill.sh skills/docling/run_docling.sh convert \
  --source "/path/to/paper.pdf" \
  --to md \
  --preset scan-heavy

OCR.space fallback is available only through explicit remote upload flags:

bash ~/.codex/runtime/run_skill.sh skills/docling/run_docling.sh convert \
  --source "/path/to/paper.pdf" \
  --to md \
  --preset scan-heavy \
  --ocr-fallback ocrspace \
  --allow-remote-ocr

To test the live OCR.space adapter, run the explicit smoke command. It generates and uploads a synthetic one-page PDF, not a user document:

bash ~/.codex/runtime/run_skill.sh skills/docling/run_docling.sh ocrspace-smoke \
  --allow-remote-ocr

Docling config can be passed with --config, AAS_DOCLING_CONFIG, DOCLING_CONFIG, or $AAS_RUNTIME_WORKSPACE/config/docling.toml. The runtime skill directory includes docling.example.toml; live config files and OCR.space keys stay outside the managed manifest.

Profiles

Profiles are named presets for installing groups of related skills. Use a profile when you want a workflow bundle instead of selecting every skill by hand. For example, research-core installs the normal research planning/source/review/delivery path, while full-research selects every canonical research skill.

make plan ARGS="--profile research-core"
make install ARGS="--profile research-core --dry-run"
ProfileDescriptionSkills
digestTracked-topic and RSS digest workflows.research-digest-wrapper, rss-news-digest, digest-bridge
documentDocument conversion and structured database lookup.docling, database-lookup
ebookEbook discovery and library handoff.calibre, vnthuquan
figureStructural figure generation and checking.tikz-draw
formal-researchOptional Lean formalization lane with local checks and declaration-search setup.formal-skeleton-helper, lean-formalization-intake, lean-explore-mcp, lean-strict-verification-gate
formal-research-remoteOptional remote formal lane setup for LeanExplore and AXLE MCP plus local formal-lane skills; install is inert and does not start remote services.formal-skeleton-helper, lean-formalization-intake, lean-explore-mcp, lean-strict-verification-gate, axiom-axle-mcp
full-researchAll research-related skills.autonomous-research-loop, autonomous-research-loop-runtime, deep-research-workflow, source-research, research-briefing, research-report-reviewer, research-verification-gate, draft-writing, zotero, calibre, getscipapers-requester, paper-lookup, submission-venue-selector, database-lookup, docling, get-available-resources, formal-skeleton-helper, lean-formalization-intake, lean-explore-mcp, lean-strict-verification-gate, axiom-axle-mcp, model-router, workspace-rearranger, research-digest-wrapper, rss-news-digest, digest-bridge, tikz-draw, sagemath, graph-verifier, agent-group-discuss, prose, cross-agent-delegation, modal-research-compute, paper-review, annotated-review, vnthuquan, self-improving-agent, session-logs, intent-interview, decision-doubt-loop, source-grounded-decisions, adversarial-boundary-gate, behavior-preserving-cleanup
libraryPaper and ebook library workflows.zotero, calibre, getscipapers-requester, paper-lookup
mathMath and graph verification workflows.sagemath, graph-verifier, formal-skeleton-helper
multi-agentMulti-agent and structured workflow orchestration.agent-group-discuss, prose, model-router, cross-agent-delegation, autonomous-research-loop, autonomous-research-loop-runtime, decision-doubt-loop
research-coreDefault research planning, source gathering, report review, and delivery verification.research-briefing, autonomous-research-loop, autonomous-research-loop-runtime, deep-research-workflow, source-research, research-report-reviewer, research-verification-gate, intent-interview
serious-researchSource-preserving research workflow with local libraries, document parsing, validation, multi-agent orchestration, and Lean skeleton support only; use formal-research for Lean verification.research-briefing, autonomous-research-loop, autonomous-research-loop-runtime, deep-research-workflow, source-research, research-report-reviewer, research-verification-gate, zotero, calibre, getscipapers-requester, paper-lookup, submission-venue-selector, docling, database-lookup, paper-review, agent-group-discuss, prose, model-router, cross-agent-delegation, get-available-resources, formal-skeleton-helper, workspace-rearranger, intent-interview, decision-doubt-loop, source-grounded-decisions, adversarial-boundary-gate, behavior-preserving-cleanup
workflow-toolsReusable planning helpers for resources, model routing, formal skeletons, and workspace organization.autonomous-research-loop, autonomous-research-loop-runtime, get-available-resources, model-router, formal-skeleton-helper, workspace-rearranger
writing-workflowClaim-preserving draft writing, rewriting, and revision-audit workflow.draft-writing

Artifact Profiles

Artifact profiles install optional support files outside normal skill folders: templates, instruction docs, reviewer personas, entrypoint aliases, and repo-management notices. They are opt-in because these files can affect agent behavior more broadly than a single skill.

make plan ARGS="--no-skills --artifact-profile workflow-templates"
make plan ARGS="--no-skills --artifact-profile repo-management"

Use --with-deps when selected artifacts should also bring in the backing skills they depend on.

Artifact ProfileDescriptionArtifacts
cross-provider-delegationTemplates and guidance for true cross-provider delegation runs.instruction-doc:cross-provider-delegation, template:cross-provider-research-panel, template:manager-worker-research-review, template:repo-comparison-research, template:evidence-synthesis-critique
repo-managementTop-level managed notice blocks for agent instruction files.management-notice:repo-management
research-entrypointsOptional command or quick-action aliases that point to backing skills.entrypoint-alias:deep-research, entrypoint-alias:research-team, entrypoint-alias:review, entrypoint-alias:tikz, entrypoint-alias:sage, entrypoint-alias:zotero, entrypoint-alias:docling, entrypoint-alias:calibre, entrypoint-alias:vnthuquan, entrypoint-alias:research-compute, entrypoint-alias:rss, entrypoint-alias:digest, entrypoint-alias:getscipapers
review-personasReviewer and research role personas rendered to each agent's supported format.agent-persona:literature-scout, agent-persona:math-explorer, agent-persona:proof-checker, agent-persona:paper-reviewer, agent-persona:code-reviewer, agent-persona:test-reviewer, agent-persona:security-reviewer
serious-researchTemplates and guidance for source-preserving, validated research runs.template:research-scope-brief, template:research-evidence-matrix, template:research-verification-checklist, template:research-workflow-runbook, template:evidence-synthesis-critique, template:deep-research-sources, template:deep-research-analysis, template:deep-research-report, template:cross-provider-research-panel, template:manager-worker-research-review, instruction-doc:research-quick-actions, instruction-doc:cross-provider-delegation
workflow-artifactsAll portable templates, workflow docs, personas, and entrypoint aliases.template:spec, template:tasks-plan, template:tasks-todo, template:draft-claim-ledger, template:draft-revision-map, template:research-scope-brief, template:research-evidence-matrix, template:research-verification-checklist, template:research-workflow-runbook, template:hierarchical-agent-delegation, template:cross-provider-research-panel, template:manager-worker-research-review, template:repo-comparison-research, template:evidence-synthesis-critique, template:deep-research-sources, template:deep-research-analysis, template:deep-research-report, instruction-doc:engineering-lifecycle, instruction-doc:operating-discipline, instruction-doc:delivery-verification-gate, instruction-doc:failure-recovery-discipline, instruction-doc:context-discipline, instruction-doc:claim-preserving-writing, instruction-doc:research-quick-actions, instruction-doc:cross-provider-delegation, instruction-doc:python-quality-gates, instruction-doc:modal-offload-routing, instruction-doc:github-actions-offload-routing, instruction-doc:scrapling-integration, instruction-doc:language-style-rules, agent-persona:literature-scout, agent-persona:math-explorer, agent-persona:proof-checker, agent-persona:paper-reviewer, agent-persona:code-reviewer, agent-persona:test-reviewer, agent-persona:security-reviewer, entrypoint-alias:deep-research, entrypoint-alias:research-team, entrypoint-alias:review, entrypoint-alias:tikz, entrypoint-alias:sage, entrypoint-alias:zotero, entrypoint-alias:docling, entrypoint-alias:calibre, entrypoint-alias:vnthuquan, entrypoint-alias:research-compute, entrypoint-alias:rss, entrypoint-alias:digest, entrypoint-alias:getscipapers, management-notice:repo-management
workflow-instructionsAgent-readable workflow guidance documents copied outside skill folders.instruction-doc:engineering-lifecycle, instruction-doc:operating-discipline, instruction-doc:delivery-verification-gate, instruction-doc:failure-recovery-discipline, instruction-doc:context-discipline, instruction-doc:claim-preserving-writing, instruction-doc:research-quick-actions, instruction-doc:cross-provider-delegation, instruction-doc:python-quality-gates, instruction-doc:modal-offload-routing, instruction-doc:github-actions-offload-routing, instruction-doc:scrapling-integration, instruction-doc:language-style-rules
workflow-templatesReusable research, specification, and task templates.template:spec, template:tasks-plan, template:tasks-todo, template:draft-claim-ledger, template:draft-revision-map, template:research-scope-brief, template:research-evidence-matrix, template:research-verification-checklist, template:research-workflow-runbook, template:hierarchical-agent-delegation, template:cross-provider-research-panel, template:manager-worker-research-review, template:repo-comparison-research, template:evidence-synthesis-critique, template:deep-research-sources, template:deep-research-analysis, template:deep-research-report
writing-workflowClaim-preserving writing workflow instructions and templates.instruction-doc:claim-preserving-writing, template:draft-claim-ledger, template:draft-revision-map

Skills

Skills are the installable agent capabilities. Installing a skill creates the per-agent SKILL.md target, support files when needed, and managed instruction blocks only for installed, adopted, or migrated skills. By default those skill targets follow auto mode: Claude links to canonical/skills, Codex and DeepSeek receive reference adapters, OpenCode receives copied native skill files plus support files, and Antigravity receives flat global Markdown adapters plus native plugin/config scaffolds unless native loader evidence justifies a different policy. Explicit symlink, reference, and copy modes force the same strategy for every agent. Use --skill or --skills for narrow installs.

make plan ARGS="--skill zotero"
make install ARGS="--skills zotero,docling --dry-run"
SkillDescriptionProfiles
adversarial-boundary-gatePre-delivery threat-model of trust boundaries and an abuse-case/injection check, delegating to a fresh-context security reviewer.serious-research, full-research
agent-group-discussMulti-agent discussion, review, and research orchestration.multi-agent, serious-research, full-research
annotated-reviewAnnotated paper review workflow when both annotation and review are requested.full-research
autonomous-research-loopRun bounded autonomous research iterations with evidence gates, recovery ledgers, and optional cross-agent handoffs.research-core, serious-research, workflow-tools, multi-agent, full-research
autonomous-research-loop-runtimeOffline runtime helper for autonomous research loop ledger initialization, iteration appends, validation, status, and selftest.research-core, serious-research, workflow-tools, multi-agent, full-research
axiom-axle-mcpOptional inert setup helper for AxiomMath AXLE MCP formal-proof assistance.formal-research-remote, full-research
behavior-preserving-cleanupClarity-only edit pass behind a comprehension gate with verify-after-each-change so behavior stays fixed.serious-research, full-research
calibreCalibre ebook lookup and library helper workflows.library, ebook, serious-research, full-research
cross-agent-delegationCross-agent delegation packet contract for bounded parent-controlled handoffs.multi-agent, serious-research, full-research
database-lookupStructured public scientific, biomedical, regulatory, materials, and economic database lookups.document, serious-research, full-research
decision-doubt-loopIn-flight fresh-context adversarial review of a non-trivial decision before it stands.serious-research, full-research, multi-agent
deep-research-workflowPhased source-preserving research workflow: search, analyze, write, with citation handoff.research-core, serious-research, full-research
digest-bridgeConvert digest output into paper retrieval manifests.digest, full-research
doclingParse, convert, OCR, chunk, and analyze documents.document, serious-research, full-research
draft-writingClaim-preserving draft writing workflow for controlled rewriting, polishing, and revision audits.writing-workflow, full-research
formal-skeleton-helperGenerate minimal Lean-style theorem skeletons, namespace wrappers, and formal statement stubs.workflow-tools, math, formal-research, formal-research-remote, serious-research, full-research
get-available-resourcesDetect CPU, memory, disk, and optional accelerator availability before heavy local work.workflow-tools, serious-research, full-research
getscipapers-requesterExternal paper retrieval fallback after local library checks.library, serious-research, full-research
graph-verifierLightweight graph sanity checks.math, full-research
intent-interviewElicit and confirm real intent one question at a time before any brief, spec, or code.research-core, serious-research, full-research
lean-explore-mcpOptional inert LeanExplore MCP setup helper for Lean declaration search.formal-research, formal-research-remote, full-research
lean-formalization-intakeOptional local-first Lean formalization intake and suitability decision workflow.formal-research, formal-research-remote, full-research
lean-strict-verification-gateScanner-first Lean artifact verification gate that separates typecheck status from claim support.formal-research, formal-research-remote, full-research
modal-research-computeRoute heavy compute jobs to Modal through a local broker.full-research
model-routerChoose an appropriate model, reasoning level, and role for subagents or multi-agent research work.workflow-tools, multi-agent, serious-research, full-research
paper-lookupExternal paper metadata and discovery fallback.library, serious-research, full-research
paper-reviewSingle-agent paper review workflow.serious-research, full-research
proseStructured reproducible research and workflow orchestration.multi-agent, serious-research, full-research
research-briefingScope nontrivial research before execution with evidence plan and workflow recommendation.research-core, serious-research, full-research
research-digest-wrapperRun tracked-topic research digests.digest, full-research
research-report-reviewerReview draft research reports for unsupported claims, ambiguity, and evidence gaps.research-core, serious-research, full-research
research-verification-gateFinal evidence, date, and gap check before delivery.research-core, serious-research, full-research
rss-news-digestRun and manage RSS digest workflows.digest, full-research
sagemathSage-backed math, graph theory, algebra, and verification.math, full-research
self-improving-agentLog durable learnings and propose canonical repo integration plans across install targets.full-research
session-logsSearch prior local agent session logs when explicitly requested.full-research
source-grounded-decisionsGround version- and spec-sensitive decisions in cited authoritative sources; flag when unverified.serious-research, full-research
source-researchGeneral web and source-gathering research workflow for current-information synthesis.research-core, serious-research, full-research
submission-venue-selectorEvidence-gated journal and conference venue selection for scholarly drafts; deliverable rankings require comparator-paper evidence.serious-research, full-research
tikz-drawStructural TikZ figure generation, compile, review, and semantic checks.figure, full-research
vnthuquanVietnam Thu Quan ebook discovery, validation, dry-run download, and Calibre dry-run handoff.ebook, full-research
vnu-eofficeRoute VNU eOffice requests to an existing vnu_eoffice package or CLI: monitor updates, list latest incoming/outgoing documents, search by keyword, download attachments, and send requested files through Telegram.
workspace-rearrangerPlan safe workspace organization with dry-run first, explicit apply, and no silent deletion.workflow-tools, serious-research, full-research
zoteroZotero paper search, retrieval, ingest, and collection workflow.library, serious-research, full-research
Ver o original no GitHub

Bom para

Relacionados