Capture and review operator feedback, frustration about agent behavior, or workflow failures when refining AGENTS.md, skills, or agent workflows. Compare relevant prior evidence before proposing changes.
Langue du texte source : anglais
Copiez ce prompt dans l’assistant IA que vous utilisez.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/ilude/dotfiles.Chargement du README du dépôt…
Affichage de 40 skills collectés sur 105.
Capture and review operator feedback, frustration about agent behavior, or workflow failures when refining AGENTS.md, skills, or agent workflows. Compare relevant prior evidence before proposing changes.
Langue du texte source : anglais
Inspect and operate ZeroOne OpenStack resources using the local cloud configuration. Use for flavors, quotas, worker capacity, networking, floating IPs, and OpenTofu deployment questions.
Langue du texte source : anglais
Local browser automation via Brave CDP. Use for logged-in checks, exact page targets, screenshots, or controlled browser comparisons.
Langue du texte source : anglais
Use native Herdr for explicitly directed communication with independent sessions on the same connected Herdr server, requested long-running processes and logs inside Herdr, or visible-agent recovery when owned subagent controls cannot reach a pane.
Langue du texte source : anglais
Review code changes for defects and relevant concerns, with evidence and priorities calibrated to the project's actual use. Use for reviews of working-tree changes, commits, branches, or pull requests, including adversarial review.
Langue du texte source : anglais
Classify and analyze tool-call and subagent failures without confusing legitimate command, test, HTTP, or application outcomes with failures of the tool mechanism. Use for session reviews, failure counts, recurring-error analysis, and diagnosis of tool or…
Langue du texte source : anglais
Analyze existing Pi sessions and usage records across default, legacy, or both profiles. Prefer log_analytics; use read-only JSONL shell tools when it cannot answer. Use for session discovery, tool failures, content lookup, and local usage analysis, not new…
Langue du texte source : anglais
Safely integrate plan or subagent branches, especially when the destination checkout has existing changes. Use for Git merges, dirty-worktree preservation, conflict handling, and post-merge cleanup.
Langue du texte source : anglais
Close out an authorized, already-committed plan task by integrating its task branch locally, recording completion metadata, and cleaning its worktree
Langue du texte source : anglais
Create, review, or explicitly revise standalone implementation plans in .specs.
Langue du texte source : anglais
Implement or review Pi extensions, tools, hooks, commands, session lifecycle, UI, subprocesses, and extension-owned prompt context or caching.
Langue du texte source : anglais
Create, review, or optimize agent instructions, including role/system prompts, tool guidance, AGENTS.md, skills, and prompt templates. Use for prompt composition, duplication, clarity, and context efficiency.
Langue du texte source : anglais
Use deferred Onclave vault tools for YouTube transcript and content workflows while keeping local fetching explicit.
Langue du texte source : anglais
Create or revise concise agent skills after the operator has decided a skill is needed. Use to select, structure, and write high-value skill instructions.
Langue du texte source : anglais
Write or review TypeScript and JavaScript, especially typed object construction, runtime boundaries, modules, async resources, package configuration, and tests.
Langue du texte source : anglais
Safely crop, resize, rotate, auto-orient, convert, or compress local images.
Langue du texte source : anglais
Guide test design when adding, changing, or reviewing tests, especially mock boundaries and test infrastructure. Not a requirement to audit existing tests or add a separate testing phase.
Langue du texte source : anglais
Analysis, diagnosis, adversarial critique, or coding mechanism selection, including routine edits. Not for diff review, architecture design, planning, or routine edit consistency.
Langue du texte source : anglais
Ansible YAML, playbooks, inventory, roles, handlers, ansible-playbook, ansible-lint, Molecule, or Galaxy.
Langue du texte source : anglais
API design, REST, GraphQL, or authentication patterns.
Langue du texte source : anglais
Module, interface, seam, dependency, or structural design. Not for approach selection, planning, or routine edit consistency.
Langue du texte source : anglais
Managed background terminals, bg_start/bg_kill, /ps, dev servers, watchers, or concurrent long-running shell commands. Not for timers or durable tasks.
Langue du texte source : anglais
Generate approaches before implementation for design decisions, architecture choices, or multiple valid solutions.
Langue du texte source : anglais
Export, archive, save, or recover Brave tabs or Chromium session files.
Langue du texte source : anglais
Local browser automation via Brave CDP. Use for logged-in checks, exact page targets, screenshots, or controlled browser comparisons.
Langue du texte source : anglais
Code, PR, diff, branch, or commit review; git diff, git show, or change comparison. Not for debugging or design critique.
Langue du texte source : anglais
C#/.NET, .cs, .csproj, .sln, dotnet, NuGet, EF, LINQ, DI, async/await, xUnit, or NUnit.
Langue du texte source : anglais
SQL, schemas, migrations, ORMs, query optimization, indexing, transactions, EXPLAIN, PostgreSQL, MySQL, SQLite, DynamoDB, or Redis.
Langue du texte source : anglais
Dockerfile, docker-compose.yml, docker-compose, .dockerignore, .devcontainer, devcontainer.json, containers, images, orchestration, or Podman.
Langue du texte source : anglais
README, CHANGELOG, docs/, RFCs, ADRs, guides, or Markdown structure. Not for prose cleanup or archival work.
Langue du texte source : anglais
Domain-Driven Design (DDD) with Eric Evans concepts: ubiquitous language, concrete domain scenarios, invariants, entity identity and lifecycle, bounded contexts, aggregates, commands, domain events, or context maps. Use when clarifying business behavior or…
Langue du texte source : anglais
Forgejo Actions: .forgejo/workflows, forgejo-runner, labels, secrets, workflow_dispatch, CI/CD, or deployment. Not for Git remotes, repository setup, or Forgejo APIs; use git-workflow.
Langue du texte source : anglais
Git operations, commits, branches, remotes, GitHub/GitLab/Forgejo repository setup, merge, rebase, reset, or history rewrite. Not for SSH client/server troubleshooting; use ssh. Not for Forgejo Actions; use forgejo-actions.
Langue du texte source : anglais
Go: .go, go.mod, go.sum, goroutines, channels, interfaces, go test, go vet, or concurrency.
Langue du texte source : anglais
Stress-test a plan or design, get grilled on a design, or "grill me".
Langue du texte source : anglais
Control Herdr terminal workspaces, tabs, panes, commands, and coding agents when the user explicitly requests Herdr. Requires HERDR_ENV=1.
Langue du texte source : anglais
Inspect and safely crop, resize, rotate, auto-orient, convert, or compress local images.
Langue du texte source : anglais
justfile/Justfile, recipes, dotenv, recursive just, or cross-platform just. Not for workflow UX or shell scripting.
Langue du texte source : anglais
Kubernetes manifests, Helm charts, Kustomize, Argo CD, kubectl, helm, GitOps, or cluster rollout checks. Not for Docker-only files.
Langue du texte source : anglais
Focused, pattern-matching edits to existing code. Not for architecture strategy or pure git operations.
Langue du texte source : anglais
Fichier source du README : README.md
Lire le README source