Skip to content
Back to skills

Paths

ASecurity

Resolves all configured document-discovery paths for the current

  • 34 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 10, 2026
ai-agentsbash

Works with

  • claude code

Security analysis

A100/100

Scanned September 10, 2026

npx -y skills add atomicinnovation/accelerator --skill paths --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Paths?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Paths
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/atomicinnovation-paths/badge)](https://www.skillsdirectory.com/skills/atomicinnovation-paths)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: paths
description: Resolves all configured document-discovery paths for the current
  project. Preloaded by agent definitions that need config-driven directory
  locations; not intended for direct user invocation.
user-invocable: false
allowed-tools: 
  - Bash(accelerator config *)
---

<!--
Maintainer note: this skill uses `user-invocable: false` (hide from the /
menu) rather than `disable-model-invocation: true`, because the latter blocks
preload via subagent `skills:` frontmatter (per Claude Code docs). Do not
change to disable-model-invocation without re-reading the subagents docs.
-->

These paths are authoritative for all document searches in this project. If
a key is missing from the resolved list below (e.g. bang preprocessing was
disabled or failed), fall back to the plugin default shown in the legend.

!`accelerator config paths --fail-safe`

## Path legend

What lives at each path key, with the plugin default if no override is set:

- `plans` — implementation plans for specific work items (default: `meta/plans`)
- `research_codebase` — codebase research outputs on specific work items (default: `meta/research/codebase`)
- `research_issues` — issue / RCA research outputs (default: `meta/research/issues`)
- `research_design_inventories` — design-inventory artifacts, one directory per snapshot (default: `meta/research/design-inventories`)
- `research_design_gaps` — design-gap analysis artifacts (default: `meta/research/design-gaps`)
- `decisions` — architectural decision records for the codebase (default: `meta/decisions`)
- `prs` — PR descriptions for landed changes (default: `meta/prs`)
- `validations` — plan validation reports (default: `meta/validations`)
- `review_plans` — reviews of implementation plans (default: `meta/reviews/plans`)
- `review_prs` — reviews of PR descriptions (default: `meta/reviews/prs`)
- `review_work` — reviews of work items (default: `meta/reviews/work`)
- `work` — work items, often `NNNN-title.md` (default: `meta/work`)
- `notes` — meeting notes, discussions, ad-hoc context (default: `meta/notes`)
- `global` — cross-repo / org-wide information (default: `meta/global`)

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.