What is pkb-agent-first?

A task manager built on PocketBase, a local backend database, with a Node worker and an MCP connection. It lets you define development tasks, send them to GLM sub-agents, verify them with tests, deploy the results, and receive updates.

Use it to create, launch, monitor, verify, and deploy development tasks through the local PocketBase and worker services.

It keeps task planning, execution, checking, and notifications in one workflow. Deterministic rules handle routine steps, while agents are used when a task needs judgment.

pkb-agent-first — Skill for Claude Code and Codex by MauricioPerera

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add MauricioPerera/KDD --skill pkb-agent-first

Clone the repo

git clone --depth 1 https://github.com/MauricioPerera/KDD

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/mauricioperera/kdd/pkb-agent-first/github.svg)](https://agentmods.dev/skills/mauricioperera/kdd/pkb-agent-first)

Your own site

<a href="https://agentmods.dev/skills/mauricioperera/kdd/pkb-agent-first"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pkb-agent-first/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/mauricioperera/kdd/pkb-agent-first"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pkb-agent-first.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,185 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00000	$0.03185
Opus 5	$0.00000	$0.01592
Sonnet 5	$0.00000	$0.00637
Haiku 4.5	$0.00000	$0.00318

Measured 12d ago against content hash d5befc5752db, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

pkb-agent-first scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://127.0.0.1:8090/api/health # API is healthy

skills/pkb-agent-first/SKILL.md · 154 lines

How it starts

The opening of the file, as written. The whole thing — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PocketBase Agent-First — guía de uso

Backend local en D:\Repo\pkb. Doc canónica completa: D:\Repo\pkb\README.md. Dos planos: determinista (reglas/hooks en PocketBase, 0 tokens) y agente (sub-agentes GLM solo cuando hace falta juicio). Idea del task-manager: defines el plan, el sistema lo ejecuta, verifica, despliega y te avisa; tú solo revisas el resultado.

0. Componentes y salud (comprueba SIEMPRE primero)

PocketBase http://127.0.0.1:8090 (binario en C:\Users\Administrador\Desktop\pocketbase_0.39.5_windows_amd64).
Worker Node http://127.0.0.1:8787 (D:\Repo\pkb\pb-worker), lanza GLM (ollama launch claude --model glm-5.2:cloud).
MCP pocketbase (D:\Repo\pkb\pb-mcp, 13 tools) — control desde una sesión de Claude.
Colecciones: agent_tasks (backbone determinista) y dev_tasks (gestor de dev-tasks).
Credenciales: superuser [email protected]; la contraseña vive SOLO en pb-worker/.env (gitignoreado). NUNCA en código ni en commits.

curl -s http://127.0.0.1:8090/api/health   # API is healthy
curl -s http://127.0.0.1:8787/health        # {"ok":true,"mode":"real",...}

1. Operación de servicios (Windows, DETACHED — sobreviven al cierre de shells)

# PocketBase (si esta caido; a veces otro proceso ocupa :8090)
$d="C:\Users\Administrador\Desktop\pocketbase_0.39.5_windows_amd64"
Start-Process -FilePath (Join-Path $d "pocketbase.exe") -ArgumentList "serve","--dev","--http=127.0.0.1:8090" `
  -WorkingDirectory $d -WindowStyle Hidden -RedirectStandardOutput (Join-Path $d "pb.out.log") -RedirectStandardError (Join-Path $d "pb.err.log")
# Worker (carga .env, modo real)
powershell -ExecutionPolicy Bypass -File D:\Repo\pkb\pb-worker\start-worker.ps1 -ExecMode real -BlockedVerify on
powershell -ExecutionPolicy Bypass -File D:\Repo\pkb\pb-worker\stop-worker.ps1

Los hooks (pb_hooks/*.pb.js) NO tienen hot-reload fiable → reiniciar pocketbase.exe tras cambiarlos.

2. Delegar una dev-task (flujo principal)

Defines la tarea CON sus tests antes que el código; al marcarla ready, el hook dispara el webhook al worker, que lanza GLM y verifica. En modo real reintenta hasta MAX_ATTEMPTS=2 si failed.

Read the full file on GitHub · 154 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

12d ago First seen · 154 lines · 0 tokens per session scan A d5befc5752db

Subscribe to this mod's changes

pkb-agent-first is a skill published in the GitHub repository MauricioPerera/KDD (8 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,185 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

LLM Project management agentic-coding ai-agents contract-driven-development developer-tools markdown

crit-cli

Use when an agent needs to author or reply to crit inline comments programmatically (including multi-agent workflows commenting on shared code/plans/docs/proposals), publish or unpublish a crit review with crit share, sync a crit review to or from a GitHub PR or GitLab MR, or read/interpret a crit review JSON file.…

tomasz-tomczyk/crit · 110 tokens

crit-story

Author a crit story and continue the interactive review loop only when the user explicitly invokes crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.

tomasz-tomczyk/crit · 47 tokens

crit

Review code changes, a plan, a live page (running dev server), or a local HTML file with Crit inline comments and structured human feedback. Use only when the user explicitly invokes /crit or directly asks to use Crit; a generic review request does not count.

tomasz-tomczyk/crit · 55 tokens

pkb-agent-first