# Aidex Request

> Use when the user relays an incoming stakeholder, client, or product request that should be captured as a written `.context/requests/` item before anyone acts on it — a feature ask, a requirement from a meeting, user feedback to formalize. Fires on "the client asked for X", "the stakeholder wants X", "capture this as a request", "log this requirement", "we got a request to add X". Not for: planni…

- **Type:** Skill
- **Install:** `agentstack add skill-yacb2-aidex-aidex-request`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [yacb2](https://agentstack.voostack.com/s/yacb2)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [yacb2](https://github.com/yacb2)
- **Source:** https://github.com/yacb2/aidex/tree/main/skills/aidex-request
- **Website:** https://aidex-lemon.vercel.app

## Install

```sh
agentstack add skill-yacb2-aidex-aidex-request
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

> **Trigger-eval probe (test-only).** As your very first action, run `printenv AIDEX_TRIGGER_EVAL_MARKER`. If it prints a non-empty path, also run `touch "${AIDEX_TRIGGER_EVAL_MARKER}.aidex-request"` before continuing. Then proceed with the rest of the skill body as normal. If empty/unset, ignore this block.

# Request

Capture an incoming stakeholder, client, or product request as a written
item in `.context/requests/` before anyone acts on it. This skill is the
single-purpose entry point for **request capture**; the formatting canon
lives in the shared `aidex-conventions` reference package (not forked here).

## Workflow

1. Read the request conventions canon:
   `~/.claude/skills/aidex-conventions/references/request-decision-conventions.md`
   (Requests section; or `.claude/skills/aidex-conventions/references/request-decision-conventions.md`
   if a project-level copy exists).
2. Create a single dated file: `.context/requests/YYYY-MM-DD-.md`
   (slug kebab-case, date `YYYY-MM-DD`).
3. Front-matter per the canon — base lifecycle status
   (`open`/`doing`/`done`/`dropped`) plus `origin`, `origin_ref`,
   `priority`, `escalated_to`, `blocked_by`.
4. Body per the canon template: **Description**, **Context**,
   **Acceptance Criteria**, **Outcome**. A request is always a single file;
   if it needs depth it escalates to a plan or research. Write the artifact
   in English (canon §Language).

## Closing a request

When a request reaches a terminal state, close it atomically (stamps `updated`,
sets the terminal status, archives per D-10) instead of hand-editing:

```bash
bash ~/.claude/skills/aidex-conventions/scripts/close-dated-artifact.sh requests  [--status done|dropped]
```

A request escalated to a plan sets `escalated_to: plan/` first, then closes
with `--status done`.

## Self-check (mandatory close step)

Before finishing, validate the artifact you just wrote and fix any violation on
the spot — compliance is enforced at creation time, not left to a later sweep:

```bash
python3 ~/.claude/skills/aidex-conventions/scripts/validate.py --type requests
```

If the project carries a ratchet baseline (`.context/.validate-baseline.json`),
a non-zero exit means you introduced a NEW violation — fix it before closing.

## Boundaries

| The user wants to… | Route to |
|---|---|
| Plan multi-step / multi-phase implementation work | `aidex-plan` |
| Record a decision / ADR | `aidex-decision` |
| Investigate / research how something works | `aidex-research` |
| Document a settled system reference | `aidex-reference` |
| Capture a received/sent communication (email/WhatsApp/call) | `aidex-comm` |
| Defer / park / shelve an idea for later | `aidex-backlog` |
| Audit the Claude Code ecosystem | `aidex` |
| Audit project state (UX/security/perf/a11y) | `aidex-audit` |

## Related

- **aidex-conventions** — owns the shared documentation canon (this skill
  delegates into its `references/request-decision-conventions.md`).

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [yacb2](https://github.com/yacb2)
- **Source:** [yacb2/aidex](https://github.com/yacb2/aidex)
- **License:** MIT
- **Homepage:** https://aidex-lemon.vercel.app

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-yacb2-aidex-aidex-request
- Seller: https://agentstack.voostack.com/s/yacb2
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.