Agent skill

Feature Spec

by PackmindHub in PackmindHub/packmind

Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description.

Apache-2.0Auto-check passedProduct & Project Management

Install Feature Spec

skills CLI
$ npx skills add PackmindHub/packmind --skill feature-spec -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install PackmindHub/packmind feature-spec --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/PackmindHub/packmind.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/feature-spec .claude/skills/feature-spec && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
feature-spec
GitHub stars
317
Token cost
~1.8k tokens
SKILL.md length
725 words
Files
8 (incl. references)
Skills in repo
35
Repo updated
First seen
Licence
Apache-2.0

At a glance

Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description.

  • Works in 5 steps: Check which output files exist to… → Read the main spec file's YAML… → If resuming, display → …
  • Tasks that involve PRD writing
  • SKILL.md covers When to Use, Packmind Repo Layout…, Input Sources and Workflow Phases, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Feature Spec is an agent skill from PackmindHub/packmind. Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description. Breaks Phase 3 into 4 fine-grained approval gates (domain data structures, ports/use cases/routes, frontend components, implementation plan) to catch problems early. Routes work between the OSS sibling (../packmind) and the proprietary repo (cwd). Outputs spec artifacts under tmp/feature-specs/{slug}/ for use with /feature-sprint.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including reference files (for example `phase-1-resolve-source.md`, `phase-2-discovery-and-spec.md` and `phase-3-implementation.md`).

It sits in Product & Project Management, covering PRD writing and Planning. It works with GitHub. The repository describes itself as: Packmind seamlessly captures your engineering playbook and turns it into AI context, guardrails, and governance. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve PRD writing
  • Tasks that involve Planning

Example prompts

  • “/feature-spec”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Check which output files exist to determine current phase (see Phase Detection table above)
  2. Read the main spec file's YAML frontmatter for status: DRAFT|COMPLETE
  3. If resuming, display
  4. If user confirms, proceed to the appropriate phase file
  5. If no task directory found, start a new feature spec

What it can do on your machine

Read from SKILL.md and the folder at commit 858ed50. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Feature Spec loads about 1.8k tokens when it runs, and up to ~6.5k if it reads all its reference files. Until then it costs about 111 tokens; SKILL.md has 725 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~111
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~6.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from PackmindHub/packmind at commit 858ed50, republished under its Apache-2.0 licence (© PackmindHub). 725 words, ~1,784 tokens.

Download SKILL.mdSave it as .claude/skills/feature-spec/SKILL.md (or your agent's skills folder). This skill also uses 7 other files; get the full folder from GitHub.
name
feature-spec
description
Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description. Breaks Phase 3 into 4 fine-grained approval gates (domain data structures, ports/use cases/routes, frontend components, implementation plan) to catch problems early. Routes work between the OSS sibling (../packmind) and the proprietary repo (cwd). Outputs spec artifacts under tmp/feature-specs/{slug}/ for use with /feature-sprint.

Feature Spec Skill

Generate a comprehensive feature specification grounded in Packmind's hexagonal architecture, frontend gateway/PM-UI conventions, and the OSS/proprietary fork boundary.

When to Use

Use this skill when the user wants to:

  • Plan a new feature, fix, or refactor that spans Packmind's stack
  • Convert a GitHub issue, design doc, or rough idea into a structured spec
  • Get fine-grained validation of data structures, routes, and components before committing to a plan

Do NOT use for trivial one-line fixes or pure dependency bumps.

Packmind Repo Layout (essential context)

  • Closed-source fork (packmind-proprietary): the current working directory when this skill runs.
  • Open-source repo (packmind): the sibling directory at ../packmind (cloned next to the proprietary repo). Resolve to an absolute path at runtime with realpath ../packmind.
  • Most feature work happens on the OSS repo and auto-merges into the proprietary fork. The proprietary fork only contains paid/closed features (e.g. packages/editions, certain packages/deployments extensions). After an OSS merge, you typically pull on the proprietary side.
  • Architecture: hexagonal (packages/{domain}/{domain,application,infra}), NestJS API at apps/api, React+@packmind/ui frontend at apps/frontend, TypeORM, Jest+@swc/jest.

Input Sources

SourceExamplesDetection
GitHub#123, owner/repo#123, full issue URLPattern #\d+ or github.com/.../issues/
Filefile:spec.md, path/to/spec.mdPrefix file: or readable file
URLhttps://...HTTP(S) URL (non-GitHub)
PromptAny other textDefault

Workflow Phases

Execute phases sequentially. Phases 1 and 4 run seamlessly (no approval gate). Phases 2 and 3 require user approval before proceeding.

PhaseFilePurposeGate
1phase-1-resolve-source.mdDetect source type, fetch content, decide OSS-vs-proprietary routing, create stateseamless
2phase-2-discovery-and-spec.mdResearch Packmind patterns (reference domain, hex layers, frontend conventions), draft acceptance criteria, generate functional specapproval
3phase-3-implementation.mdTechnical approach + implementation plan (4 subtasks below)4 approvals
3A↳ Subtask 3AValidate domain data structures (entities, value objects, events, migrations)approval
3B↳ Subtask 3BValidate ports, contracts, use cases, services, adapters, NestJS routesapproval
3C↳ Subtask 3CValidate frontend components, gateways, query hooks, PM-UI usageapproval
3D↳ Subtask 3DGenerate full implementation plan (task breakdown, grouping, coverage)approval
4phase-4-finalize.mdGenerate context.md, mark spec complete, report (no commit — artifacts live in tmp/)seamless

State Management

State is persisted to markdown files under a git-ignored tmp/ directory for cross-session continuity. Phase progress is inferred from which output files exist — no separate state file needed.

tmp/feature-specs/{slug}/
├── {slug}.md               # Main spec (status: DRAFT → COMPLETE in frontmatter)
├── discovery.md            # Phase 2 output (YAML frontmatter + markdown)
├── functional-spec.md      # Phase 2 output (informed by discovery)
├── implementation-plan.md  # Phase 3 output (includes parallel groups section)
└── context.md              # Phase 4 output (YAML frontmatter + markdown)
Phase Detection (Resume Logic)
Files PresentCompleted PhaseResume At
{slug}.md onlyPhase 1Phase 2
+ discovery.md, functional-spec.mdPhase 2Phase 3
+ implementation-plan.mdPhase 3Phase 4
+ context.md (status: COMPLETE)Phase 4Done

Invocation

Auto-Detection (Resuming)

Before starting, check for an existing task directory at tmp/feature-specs/{slug}/:

  1. Check which output files exist to determine current phase (see Phase Detection table above)
  2. Read the main spec file's YAML frontmatter for status: DRAFT|COMPLETE
  3. If resuming, display:
    **Resuming Feature Spec**: {slug}
    **Current Phase**: {detected_phase}
    **Status**: {status from frontmatter}
    
    Continue from Phase {detected_phase}? (Y/n)
  4. If user confirms, proceed to the appropriate phase file
  5. If no task directory found, start a new feature spec
Show full SKILL.md (260 more words)Show less
Starting New

To start a new feature spec: Read phase-1-resolve-source.md and follow its instructions.

<feature-spec-rules>
  <rule>Execute phases SEQUENTIALLY — never skip phases</rule>
  <rule>Phases 1 and 4 proceed automatically — no approval gate</rule>
  <rule>Phases 2 and 3 require USER APPROVAL via AskUserQuestion before proceeding</rule>
  <rule>Save all artifacts under tmp/feature-specs/{slug}/ — never under tracked folders</rule>
  <rule>Use Task tool with Explore subagent for codebase research; never read 50 files yourself</rule>
  <rule>Agent NEVER implements code in this skill — output ONLY specification documents</rule>
  <rule>Main spec file has status DRAFT until Phase 4 completes</rule>
  <rule>Always record target_repo (oss | proprietary | both) decided in Phase 1; consumed by /feature-sprint</rule>
  <rule>Never invent OSS/proprietary boundaries — if unsure, ask the user</rule>
</feature-spec-rules>

Output

Final deliverable: tmp/feature-specs/{slug}/{slug}.md + sibling artifacts.

Contains:

  • Functional specification with acceptance criteria
  • Technical approach grounded in Packmind reference patterns
  • Implementation plan with phased tasks (each task has reference file:line patterns)
  • Parallel Groups section (for parallel execution via /feature-sprint)
  • target_repo: oss | proprietary | both

Parallel Execution Support

Phase 3 includes a grouping step that analyzes task dependencies and creates parallel execution groups:

How grouping works:

  1. After generating the implementation plan, a Plan subagent analyzes tasks
  2. Tasks are grouped by file dependencies (tasks sharing files go together)
  3. Groups are non-overlapping (no file belongs to multiple groups)
  4. Results are stored in implementation-plan.md (Parallel Groups section) and context.md (YAML frontmatter)

When grouping is skipped:

  • Tasks have no clear file dependencies
  • Single-file or trivial implementations
  • User opts out during Phase 3 review

Handoff

When the spec is complete (Phase 4), the next step is /feature-sprint {slug} — the companion skill that executes the implementation plan.

© PackmindHub, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 7 other files (references) in .agents/skills/feature-spec of PackmindHub/packmind.

  • SKILL.md
  • phase-1-resolve-source.md
  • phase-2-discovery-and-spec.md
  • phase-3-implementation.md
  • phase-4-finalize.md
  • references/context-schema.md
  • references/implementation-plan-prompt.md
  • references/spec-templates.md

Open the folder on GitHubat commit 858ed50

Compare with similar skills

Feature Spec next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Feature Spec compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Feature Spec this skillPackmindHub/packmind317—~1.8kAutomated safety check: PassApache-2.0
PRP Implementation PlannerWirasm/prp2.3k—~4.1kAutomated safety check: PassMIT
CCPM Project Managementautomazeio/ccpm8.4k—~1.1kAutomated safety check: PassMIT
Ouroboros PM InterviewQ00/ouroboros6.2k—~5.7kAutomated safety check: PassMIT
AlignmentOvid/paad131—~4.9kAutomated safety check: PassMIT
Creating Issuesopsmill/infrahub529—~1.2kAutomated safety check: PassApache-2.0

Similar skills

  • Turns a PRD, issue or description into an implementation-ready plan grounded in codebase evidence, adding root-cause analysis for bugs and publishing issue plans back to the issue.

    2.3k GitHub stars~4.1k tokensUpdated 5 days ago
    DevelopmentAuto-check passed
  • Runs a spec-driven workflow from PRD to epic to GitHub issues to parallel agents, with status, standup and blocked-work reports from bundled scripts.

    8.4k GitHub stars~1.1k tokensUpdated 6 mo ago
    Product & Project ManagementAuto-check passed
  • Runs a guided product-manager interview that classifies each question automatically and produces a Product Requirements Document.

    6.2k GitHub stars~5.7k tokensUpdated yesterday
    Product & Project ManagementAuto-check passed
  • Alignment

    Ovid/paad

    A skill your agent uses when verifying that requirements/specs/PRDs and their implementation plans match — before starting work, after a spec or plan update, or when suspecting coverage gaps, scope…

    131 GitHub stars~4.9k tokensUpdated yesterday
    Product & Project ManagementAuto-check passed
  • Creating Issues

    opsmill/infrahub

    Turns a single feature idea, improvement, or bug into ONE well-structured GitHub issue.

    529 GitHub stars~1.2k tokensUpdated today
    Product & Project ManagementAuto-check passed
  • Clarification Analyst

    GulajavaMinistudio/Mayukai-Theme

    Helps interrogate Product Requirements (PRD), Technical Specifications, and Implementation Plans to find ambiguities, missing edge cases, and hidden assumptions.

    139 GitHub stars~1.1k tokensUpdated 3 mo ago
    Product & Project ManagementAuto-check passed

More from PackmindHub/packmind

All 35 skills in this repo
  • Michel CLI Demo Recorder

    PackmindHub/packmind

    Produce proof-of-execution demos of the Packmind CLI (packmind-cli) as terminal-styled images (colors and formatting preserved exactly), for embedding in a GitHub PR.

    317 GitHub stars~3.4k tokensUpdated yesterday
    Auto-check passed
  • Michel UI Demo Recorder

    PackmindHub/packmind

    Record polished UI demo videos and screenshots of a running web app using Playwright MCP — for client deliverables, release notes, feature walkthroughs, or bug repros.

    317 GitHub stars~6.4k tokensUpdated yesterday
    Auto-check passed
  • Packmind Create Skill

    PackmindHub/packmind

    Guide for creating effective skills. An agent skill from PackmindHub/packmind.

    317 GitHub stars~3.5k tokensUpdated yesterday
    Auto-check: notes
  • Doc Audit

    PackmindHub/packmind

    Audit Packmind end-user documentation (apps/doc/) for broken links, outdated CLI references, non-existent concepts, misleading information, and missing coverage.

    317 GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed
  • Feature Sprint

    PackmindHub/packmind

    Execute the implementation plan produced by /feature-spec. An agent skill from PackmindHub/packmind.

    317 GitHub stars~1.9k tokensUpdated yesterday
    Auto-check passed
  • Review an implemented GitHub issue the way a senior Packmind engineer would — the human-judgment checks that ESLint, the TypeScript compiler, and e2e tests cannot catch (authorization scoping…

    317 GitHub stars~2.7k tokensUpdated yesterday
    Auto-check passed

Works with

Questions about Feature Spec

What does Feature Spec do?

Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description. Feature Spec is an agent skill from PackmindHub/packmind. Generate a Packmind feature specification from a GitHub issue, file, URL, or direct description.

When should I use Feature Spec?

Feature Spec fits situations like: tasks that involve PRD writing; tasks that involve Planning.

How do I install Feature Spec in Claude Code?

Run `npx skills add PackmindHub/packmind --skill feature-spec -a claude-code`. Or copy the skill folder (.agents/skills/feature-spec in PackmindHub/packmind) into .claude/skills/feature-spec in your project. Claude Code loads it when a task matches its description.

How do I install Feature Spec in Codex?

Run `npx skills add PackmindHub/packmind --skill feature-spec -a codex`. Or copy the skill folder (.agents/skills/feature-spec in PackmindHub/packmind) into .agents/skills/feature-spec in your project. Codex loads it when a task matches its description.

Can I use Feature Spec in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add PackmindHub/packmind --skill feature-spec -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/feature-spec, .gemini/skills/feature-spec, .github/skills/feature-spec and .opencode/skills/feature-spec in your project.

What does Feature Spec need to run?

SKILL.md names no scripts, command-line tools or credentials: Feature Spec is instructions for the agent only.

Does Feature Spec access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Feature Spec safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Feature Spec use?

Feature Spec is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Feature Spec use?

About 1.8k tokens (SKILL.md is roughly 7.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.7k tokens, read only when the agent opens those files.

What are the alternatives to Feature Spec?

Skills that share tags, products or a category with Feature Spec: PRP Implementation Planner (Wirasm/prp, 2.3k stars), CCPM Project Management (automazeio/ccpm, 8.4k stars), Ouroboros PM Interview (Q00/ouroboros, 6.2k stars) and Alignment (Ovid/paad, 131 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Feature Spec?

PackmindHub (a GitHub organization) maintains it in PackmindHub/packmind, which has 317 GitHub stars. The repository holds 35 skills in this directory. The repository was last updated on October 6, 2026.

Source: PackmindHub/packmind on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.