Skip to content
Agentmart
Submit a tool

using-pco

Use when starting any serious task or whenever PCO resources may apply; bootstraps deterministic cognitive routing before action.

Skill by Sigit-AP

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/sigit-ap-using-pco

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when starting any serious task or whenever PCO resources may apply; bootstraps deterministic cognitive routing before action.

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/using-pco

What's inside

Skill name
using-pco
Description
Use when starting any serious task or whenever PCO resources may apply; bootstraps deterministic cognitive routing before action.
License
MIT
Instructions
75 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: "using-pco" does not match folder "pco-using-pco"

Files (1)

  • SKILL.md

SKILL.md (first lines)

<EXTREMELY-IMPORTANT>
If there is even a 1% chance any PCO resource applies, use PCO before responding or acting.

This is not optional for planning, coding, debugging, auditing, research, architecture, reasoning, or verification.
</EXTREMELY-IMPORTANT>

## Priority

1. User instructions are highest priority.
2. PCO skills/resources define HOW to think and execute.
3. Default model habits are lowest priority.

## Mandatory Startup

PCO starts with an AMT-native boot contract, not blind copying:

1. **Amati** — detect task intent, risk, uncertainty, and applicable cognitive resources.
2. **Tiru** — reuse the proven first-use discipline: boot context first, route before action, avoid duplicate injection.
3. **Modifikasi** — make it PCO-specific: executable graph routing, evidence gates, full-read coverage, and runtime audit.

Use zero-dependency bootstrap and executable runtime first:

```bash
node scripts/pco/bootstrap.mjs --json
node references/runtime/pco-reference-runtime.mjs route "<task>" --limit 8 --depth 1
node references/runtime/pco-reference-runtime.mjs context "<task>" --limit 8 --depth 1
```

Folder-specific runtimes exist for targeted loading:

```bash
node references/core/runtime.mjs route "<task>"
node references/cognitive-engines/runtime.mjs route "<task>"
node references/quality-safety/runtime.mjs route "<task>"
node references/knowledge-bases/runtime.mjs route "<task>"
node references/advanced/runtime.mjs route "<task>"
node references/workflows/runtime.mjs route "<task>"
```

Generated `.pco/cache/*.json` files are compatibility artifacts only. Do not route from JSON when the executable runtime is available.

## Rule

Before acting:

1. Identify task type.
2. Run the reference runtime route for the task.
3. Load startup files returned by the route.
4. Load required route files and folder runtime outputs.
5. Expand via prerequisites, concepts, sections, and graph neighbors.
6. Execute with the PCO phase pipeline.
7. Verify through quality/safety gates.

## Red Flags

| Thought | Reality |
|---|---|
| "This is simple" | Simple tasks still need skill check. |
| "I remember the framework" | Runtime graph changes; run current route. |
| "I can answer first" | PCO runtime route check comes before output. |
| "I'll inspect later" | Context selection is the first step. |
| "One file is enough" | PCO is graph-based; expand by neighbors and dependencies. |

## Completion Gate

Do not finish until:

- `npm run pco:validate` passes,
- `npm run pco:healthcheck` passes for infrastructure changes,
- `npm run pco:scorecard` passes before any expanded infrastructure claim,
- evidence exists for all claims,
- quality/safety resources were considered,
- the selected route and loaded resources are known.

From Sigit-AP/Pure-Cognitive-Oil, MIT.

More from Sigit-AP/Pure-Cognitive-Oil

Signals

Stars
0
Forks
0
Open issues
1
Last commit (this folder)
Jun 27, 2026 (4 months ago)
Last release
May 28, 2026 (5 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Same skill elsewhere

Copies of this SKILL.md published by others; search shows only one of them.

More from Sigit-AP

Cite this page

using-pco skill. Agent Mart. https://agentmart.dev/skills/sigit-ap-using-pco (last updated 2026-10-24)