Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Dr Set Goal

ASecurity

Turn a vague DeepReason problem statement into one bounded, falsifiable tranche goal (GOAL.md). Use at the start of every tranche, before any diagnosis or code reading.

142 stars
0 votes
0 copies
0 views
Added 9/19/2026
testinggo

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add AHepi/DeepReason --skill dr-set-goal --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Dr Set Goal?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Dr Set Goal
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/ahepi-dr-set-goal/badge)](https://www.skillsdirectory.com/skills/ahepi-dr-set-goal)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: dr-set-goal
description: Turn a vague DeepReason problem statement into one bounded, falsifiable tranche goal (GOAL.md). Use at the start of every tranche, before any diagnosis or code reading.
---

# Set the tranche goal

Input: a problem statement (from the operator, a failed run, RESULTS.md,
or PARKED.md). Output: exactly one `GOAL.md`. Nothing else. You do not
read source code in this phase beyond confirming a file exists.

## Procedure

1. Restate the problem as ONE observable, checkable statement about the
   record or the tests. Good: "run-9175f0ec terminated budget_denied on
   all 8 question turns with zero provider calls." Bad: "the scheduler
   has issues."
2. Classify it:
   - `defect` — record/tests show behavior contradicting a documented
     guarantee (spec §, docstring, RESULTS.md claim).
   - `regression-risk` — behavior is undocumented but a change would
     alter it.
   - `capability-gap` — nothing is broken; something is missing.
   Only `defect` tranches may proceed to implementation without
   explicit operator approval; the other two stop after FIX.md
   (proposal) and report.
3. Write the success criterion as a command + expected output. It must
   be decidable by a machine, e.g.:
   - "`verify_root(<root>)` returns zero `conjecture-context`
     violations" or
   - "`pytest tests/test_x.py::test_y` passes and the full gate stays
     at 0 failed."
4. Write the boundary list: files/subsystems presumed in scope (max 3),
   and an explicit NOT-IN-SCOPE line for the nearest tempting neighbor.
5. Size check: if you cannot imagine the fix under ~150 changed lines
   and one commit, split the problem and pick the FIRST piece only;
   put the rest in PARKED.md.

## GOAL.md template (fill every field; delete nothing)

    # Goal: <one line>
    Class: defect | regression-risk | capability-gap
    Observed: <one sentence, with the record/test evidence pointer>
    Success criterion (machine-decidable):
        <command>
        <expected output>
    In scope: <max 3 paths/subsystems>
    NOT in scope: <the nearest tempting thing you will not touch>
    Budget: <=150 changed lines, 1 commit, <n> hours
    Stop conditions inherited from orchestrator: yes

## Exit criteria

- GOAL.md exists in the tranche directory, committed and pushed.
- You have NOT proposed a cause, a fix, or read implementation code.
- Return to the orchestrator.

Attribution

AHepiAHepi
View sourceMore from AHepi →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Screen Reader Testing

Practical guide to testing web applications with screen readers for comprehensive accessibility validation.

393431 votes

Python Testing

使用pytest、TDD方法、夹具、模拟、参数化和覆盖率要求的Python测试策略。

2456590 votes

Tdd Workflow

在编写新功能、修复错误或重构代码时使用此技能。强制执行测试驱动开发,包含单元测试、集成测试和端到端测试,覆盖率超过80%。

2456590 votes

Springboot Tdd

使用JUnit 5、Mockito、MockMvc、Testcontainers和JaCoCo进行Spring Boot的测试驱动开发。适用于添加功能、修复错误或重构时。

2456590 votes

Eval Harness

克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则

2456590 votes
View all in testing →