agentleFS
hmj1026hmj1026/dhpk/skills/project-brief/SKILL.md

project-brief

Skill2 starsChanged 7 days agoMITView on GitHub

Use when converting one readable technical specification into a PM or CTO executive brief. Not for editing the source specification or preserving implementation detail. Output: a saved executive-summary path and retained unresolved decisions.

10 other projects use a version of this, 2 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Simplify deliberately: Remove code snippets, internal module names, and implementation detail.

    line 43
  2. Distinctive:

    Write this Markdown shape, replacing placeholders with evidence from the source:

    line 62

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Project brief

Convert one technical specification into a PM/CTO-readable executive summary. The input path in $ARGUMENTS is read-only. Parse an optional --output <path> without changing any other argument; when omitted, save beside the source with the -brief suffix. Stop with BLOCKED when the source is missing or unreadable, or when the requested output cannot be written.

$PROJECT_DIR denotes the consumer project root in path examples. It is notation for that root, not a required ambient environment variable. The source and any --output path are consumer project files; preserve the explicit output path or the default beside-source path described above.

When NOT to Use

  • The source technical specification needs correction: edit it through its owning workflow first.
  • The audience needs implementation-level detail, code analysis, or module names: return the technical specification instead.
  • More than one source document or a broad documentation refresh is in scope: use a bounded documentation workflow.

Workflow

  1. Read the source. Resolve the path from $ARGUMENTS, record the source path, and read it completely. Extract the core value proposition. Preserve unresolved decisions instead of silently resolving them.
  2. Extract the business view. Capture one sentence covering what, why, and value; compare current state with target; retain solution pros/cons and the recommendation; keep milestones, risks, dependencies, timeline, and resources.
  3. Simplify deliberately. Remove code snippets, internal module names, and implementation detail. Keep a plain-language architecture overview with no more than three layers. Turn immediate actions into PM/CTO decision points.
  4. Write the brief. Save to the explicit --output path or to the source directory using <stem>-brief<extension>. Do not modify the source file.

The conversion map is:

Technical sectionExecutive treatment
Trust boundary diagramSimplify to at most three architecture layers
Code analysis; reusable modulesRemove
System architecture; implementation roadmapKeep in simplified form
Key design decisionsRemove implementation detail
Alternative comparison; risks and mitigations; timelineKeep
Immediate actionsSimplify to decision points

Output

Write this Markdown shape, replacing placeholders with evidence from the source:

# [Project Name] Executive Summary

## Project Overview
> One sentence: what, why, and value

## Current State vs Target
| Dimension | Current | Target |
| --------- | ------- | ------ |
|           |         |        |

## Solution Evaluation
| Solution | Pros | Cons | Recommendation |
| -------- | ---- | ---- | -------------- |
| Option A |      |      |                |
| Option B |      |      |                |
| Recommended |   |      | ✅ Adopt       |

## Architecture Overview
(Simplified system diagram, three layers maximum)

## Milestones
| Week | Deliverable | Dependencies |
| ---- | ----------- | ------------ |
| Week 1 |         |              |

## Risk Summary
| Risk | Impact Level | Mitigation |
| ---- | ------------ | ---------- |
|      |              |            |

## Resource Requirements
- **Headcount**:
- **Timeline**:
- **External Dependencies**:

## Decision Points
> Items requiring PM/CTO decision
- [ ] Decision 1:
- [ ] Decision 2:

Return the saved output path and list decisions retained for review. A brief is an applied documentation write, not evidence that the underlying proposal was approved or implemented.

Verification

  • The source was readable and remains unchanged.
  • The output exists at the requested or default path and is readable.
  • Code snippets, internal module names, and unnecessary implementation details are absent.
  • Business value, current/target state, alternatives, milestones, risks, resources, and decision points are present.
  • Unresolved decisions are retained and the report names the output path.

References

  • Native file reads and writes are sufficient; use repository search only to clarify names that the source explicitly references.
  • $PROJECT_DIR/docs/agent-guidance/command-contract.md, when present in the consumer checkout, provides the shared failure/output vocabulary; this skill keeps the required BLOCKED and output-path boundary locally.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.