Skip to content
Agentmart
Submit a tool

verify-product-discovery-coaches

Verify the Product Discovery Coaches Codex and Claude Code plugin package through its real CLI installation and validation paths.

Skill by brunogama

IndexedMIT
Write a review

Chalk Labs

Need help deploying verify-product-discovery-coaches?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/brunogama-verify-product-discovery-coaches

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

Verify the Product Discovery Coaches Codex and Claude Code plugin package through its real CLI installation and validation paths. Use after changing manifests, marketplaces, skills, or installation documentation.

Use cases: agent-building

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 and code. Checked Nov 2, 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/verify-product-discovery-coaches

What's inside

Skill name
verify-product-discovery-coaches
Description
Verify the Product Discovery Coaches Codex and Claude Code plugin package through its real CLI installation and validation paths. Use after changing manifests, marketplaces, skills, or installation documentation.
License
MIT
Instructions
124 lines in SKILL.md · view on GitHub

Files (5)

  • SKILL.md
  • README.md · features/
  • claude-code-package-validation.md · features/
  • coach-skill-catalogue.md · features/
  • codex-marketplace-installation.md · features/

SKILL.md (first lines)

# Verify Product Discovery Coaches

This repository ships a plugin package rather than a long-running application.
The real user surfaces are Codex marketplace installation and Claude Code plugin
validation. Do not start an interactive Claude or Codex session for this skill.

---

## Launch

No server or build is required. Start each verification run from the repository
root with an isolated Codex home and a durable evidence directory:

```bash
export RUN_ID="$(date -u +%Y%m%dT%H%M%SZ)"
export ARTIFACT_DIR=".artifacts/verification/$RUN_ID"
export CODEX_VERIFY_HOME="$(mktemp -d)"
mkdir -p "$ARTIFACT_DIR"
```

The package is ready to drive when `claude`, `codex`, and the repository root
manifests are available. The Codex run is isolated by `CODEX_HOME`; it does not
read or change the user's normal Codex installation state.

---

## Doctor

Before driving, run this read-only package health check:

```bash
claude plugin validate .
python3 - <<'PY'
import json
from pathlib import Path
for path in (
    'plugin.json',
    '.codex-plugin/plugin.json',
    '.claude-plugin/plugin.json',
    '.agents/plugins/marketplace.json',
    '.claude-plugin/marketplace.json',
):
    json.loads(Path(path).read_text(encoding='utf-8'))
    print(f'valid JSON: {path}')
PY
```

Drive the package only when Claude Code reports validation success and every
manifest parses as JSON. This check does not mutate a profile, install a plugin,
or call a model.

---

## Drive

Use the exact marketplace name and plugin identifier from the manifests. Drive
the Codex installation path as a user would, while writing all command output
to the evidence directory:

```bash
CODEX_HOME="$CODEX_VERIFY_HOME" \
  codex plugin marketplace add . --json \
  | tee "$ARTIFACT_DIR/codex-marketplace-add.json"
CODEX_HOME="$CODEX_VERIFY_HOME" \
  codex plugin add product-discovery-coaches@product-discovery-coaches --json \
  | tee "$ARTIFACT_DIR/codex-plugin-add.json"
CODEX_HOME="$CODEX_VERIFY_HOME" \
  codex plugin list \
  | tee "$ARTIFACT_DIR/codex-plugin-list.txt"
claude plugin validate . \
  | tee "$ARTIFACT_DIR/claude-plugin-validate.txt"
```

The proof passes only when the marketplace-add result names
`product-discovery-coaches`, the plugin-add result reports version `1.0.0`, the
Codex list reports it as `installed, enabled`, and Claude Code reports
`Validation passed`.

Use the feature map before expanding coverage. The Codex installation recipe is
the only fully isolated drive in this repository. The Claude Code feature is a
real package validator, not an interactive model execution; do not claim a
coach prompt was executed unless an isolated Claude session actually ran it.

---

## Evidence

Keep each run's output under `.artifacts/verification/$RUN_ID/`. That directory
is ignored by Git so evidence does not become product content. Capture the
commands above and their resulting state, not merely successful exit codes.

A valid proof exercises the real marketplace and installation path, captures
the add result and installed-plugin list, and captures the independent Claude
Code validation result. It must not use internal state setters, replace a
marketplace with handcrafted cache files, or rely on a test-only endpoint.

The Codex sandbox writes a cache below `$CODEX_VERIFY_HOME`. Confirm that the
plugin list identifies the package as installed before cleanup. The commands do
not use a dry-run mode, so observe the generated temporary cache rather than
assuming nothing was written.

---

## Cleanup

Remove only the isolated Codex home created in Launch:

```bash
rm -rf "$CODEX_VERIFY_HOME"
unset CODEX_VERIFY_HOME
```

Do not kill processes by name: this verification starts no persistent process.
Do not remove `$ARTIFACT_DIR`; evidence must survive cleanup. Confirm it remains
with `test -d "$ARTIFACT_DIR"`.

---

## Helpers

From brunogama/product-discovery-coaches, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 3, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from brunogama

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get verify-product-discovery-coaches running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying verify-product-discovery-coaches?

CallGet help
Cite this page

verify-product-discovery-coaches skill. Agent Mart. https://agentmart.dev/skills/brunogama-verify-product-discovery-coaches (last updated 2026-10-24)