<feed xmlns='http://www.w3.org/2005/Atom'>
<title>superpowers.git/tests, branch opencode-native-skills</title>
<subtitle>Agentic skills framework and development workflow
</subtitle>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/'/>
<entry>
<title>fix: preserve original event type, use source field for wrapper</title>
<updated>2026-01-17T21:11:53+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2026-01-17T21:11:53+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=fccb5b4b8f7b46c4fd37ddd6b9d4b82db50bf325'/>
<id>fccb5b4b8f7b46c4fd37ddd6b9d4b82db50bf325</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>test: add brainstorm server integration tests</title>
<updated>2026-01-17T21:03:04+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2026-01-17T21:03:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=7e86703081d4a605fcf97a3d10a4423c70c8cc9d'/>
<id>7e86703081d4a605fcf97a3d10a4423c70c8cc9d</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>Strengthen using-superpowers for explicit skill requests</title>
<updated>2025-12-27T04:41:22+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-12-27T04:41:22+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=3dac35e0b35c2fbe4222ba825381eb764d65f085'/>
<id>3dac35e0b35c2fbe4222ba825381eb764d65f085</id>
<content type='text'>
Addresses failure mode where Claude skips skill invocation even when
user explicitly requests it by name (e.g., "subagent-driven-development,
please").

Skill changes:
- "Check for skills" → "Invoke relevant or requested skills"
- "BEFORE ANY RESPONSE" → "BEFORE any response or action"
- Added reassurance that wrong skill invocations are okay
- New red flag: "I know what that means"

Tests:
- New test suite for explicit skill requests
- Single-turn and multi-turn scenarios with --continue
- Tests with haiku model and user CLAUDE.md
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Addresses failure mode where Claude skips skill invocation even when
user explicitly requests it by name (e.g., "subagent-driven-development,
please").

Skill changes:
- "Check for skills" → "Invoke relevant or requested skills"
- "BEFORE ANY RESPONSE" → "BEFORE any response or action"
- Added reassurance that wrong skill invocations are okay
- New red flag: "I know what that means"

Tests:
- New test suite for explicit skill requests
- Single-turn and multi-turn scenarios with --continue
- Tests with haiku model and user CLAUDE.md
</pre>
</div>
</content>
</entry>
<entry>
<title>Add skill triggering test framework</title>
<updated>2025-12-18T00:44:52+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-12-12T05:10:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=ae0ef56b442c106e4faec603bf65ba89c6c2cf39'/>
<id>ae0ef56b442c106e4faec603bf65ba89c6c2cf39</id>
<content type='text'>
Creates tests/skill-triggering/ to validate skills trigger correctly from
naive prompts (without explicitly naming the skill). Tests 6 skills:
- systematic-debugging
- test-driven-development
- writing-plans
- dispatching-parallel-agents
- executing-plans
- requesting-code-review

All 6 tests pass - skills trigger correctly based on their descriptions.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Creates tests/skill-triggering/ to validate skills trigger correctly from
naive prompts (without explicitly naming the skill). Tests 6 skills:
- systematic-debugging
- test-driven-development
- writing-plans
- dispatching-parallel-agents
- executing-plans
- requesting-code-review

All 6 tests pass - skills trigger correctly based on their descriptions.
</pre>
</div>
</content>
</entry>
<entry>
<title>Document skill description trap: descriptions override flowcharts</title>
<updated>2025-12-18T00:44:52+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-12-12T04:03:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=2a19be0b7824ade92420e496f9127e8ad7ed19ba'/>
<id>2a19be0b7824ade92420e496f9127e8ad7ed19ba</id>
<content type='text'>
Testing revealed that skill descriptions summarizing workflow cause
Claude to follow the description instead of reading the skill body.

- A description saying "code review between tasks" caused ONE review
- The flowchart clearly showed TWO reviews (spec compliance + quality)
- Minimal description ("Use when...") correctly deferred to flowchart

Updated writing-skills with:
- "Description = When to Use, NOT What the Skill Does" section
- Cautionary tale about this actual failure
- Examples of good (triggers only) vs bad (workflow summary) descriptions

Updated subagent-driven-development:
- Removed workflow summary from description
- Now just: "Use when executing implementation plans..."

Updated test runner:
- Added --dangerously-skip-permissions for automated testing
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Testing revealed that skill descriptions summarizing workflow cause
Claude to follow the description instead of reading the skill body.

- A description saying "code review between tasks" caused ONE review
- The flowchart clearly showed TWO reviews (spec compliance + quality)
- Minimal description ("Use when...") correctly deferred to flowchart

Updated writing-skills with:
- "Description = When to Use, NOT What the Skill Does" section
- Cautionary tale about this actual failure
- Examples of good (triggers only) vs bad (workflow summary) descriptions

Updated subagent-driven-development:
- Removed workflow summary from description
- Now just: "Use when executing implementation plans..."

Updated test runner:
- Added --dangerously-skip-permissions for automated testing
</pre>
</div>
</content>
</entry>
<entry>
<title>Add .claude settings to scaffolds for auto-approved permissions</title>
<updated>2025-12-18T00:44:52+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-12-11T23:55:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=cd83439bb2411b4c5300e8724578d7ca18d0d16b'/>
<id>cd83439bb2411b4c5300e8724578d7ca18d0d16b</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>Add test suite for subagent-driven-development skill</title>
<updated>2025-12-18T00:44:51+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-12-11T20:15:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=baef5241aae440b29ce30a66e65585602a722443'/>
<id>baef5241aae440b29ce30a66e65585602a722443</id>
<content type='text'>
Two test projects for validating the skill with full end-to-end runs:
- go-fractals: 10 tasks, CLI tool with Sierpinski and Mandelbrot
- svelte-todo: 12 tasks, CRUD app with localStorage and Playwright

Each test has design.md, plan.md, and scaffold.sh. Run with:
  ./tests/subagent-driven-dev/run-test.sh go-fractals
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Two test projects for validating the skill with full end-to-end runs:
- go-fractals: 10 tasks, CLI tool with Sierpinski and Mandelbrot
- svelte-todo: 12 tasks, CRUD app with localStorage and Playwright

Each test has design.md, plan.md, and scaffold.sh. Run with:
  ./tests/subagent-driven-dev/run-test.sh go-fractals
</pre>
</div>
</content>
</entry>
<entry>
<title>Add token usage analysis to subagent-driven-development test</title>
<updated>2025-12-18T00:44:50+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-11-30T04:51:18+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=faa65e7163e2fa6cfdc5b7277b38f2c2b7a8b688'/>
<id>faa65e7163e2fa6cfdc5b7277b38f2c2b7a8b688</id>
<content type='text'>
- Rewrote analyze-token-usage.py to parse main session file correctly
- Extracts usage from toolUseResult fields for each subagent
- Shows breakdown by agent with descriptions
- Integrated into test-subagent-driven-development-integration.sh
- Displays token usage automatically after each test run
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
- Rewrote analyze-token-usage.py to parse main session file correctly
- Extracts usage from toolUseResult fields for each subagent
- Shows breakdown by agent with descriptions
- Integrated into test-subagent-driven-development-integration.sh
- Displays token usage automatically after each test run
</pre>
</div>
</content>
</entry>
<entry>
<title>fix: verify skill usage via session transcript not text output</title>
<updated>2025-12-18T00:44:50+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-11-30T03:35:43+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=24ca8cd9d5fd83197ebe60ca45eb87e3476d6aa6'/>
<id>24ca8cd9d5fd83197ebe60ca45eb87e3476d6aa6</id>
<content type='text'>
The skill instructions are internal and don't appear in user-facing
output. Updated verification to parse the session JSONL transcript
and check for actual tool usage:
- Skill tool invocation
- Task tool (subagents)
- TodoWrite (tracking)
- Implementation results
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The skill instructions are internal and don't appear in user-facing
output. Updated verification to parse the session JSONL transcript
and check for actual tool usage:
- Skill tool invocation
- Task tool (subagents)
- TodoWrite (tracking)
- Implementation results
</pre>
</div>
</content>
</entry>
<entry>
<title>test: use bypassPermissions mode for unrestricted testing</title>
<updated>2025-12-18T00:44:50+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2025-11-30T01:15:21+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=8fbeca830a89027f9a40ee7cdcfee54202b9f65a'/>
<id>8fbeca830a89027f9a40ee7cdcfee54202b9f65a</id>
<content type='text'>
dontAsk mode was auto-denying Write tool. Use bypassPermissions
instead to allow full tool access in this controlled test environment.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
dontAsk mode was auto-denying Write tool. Use bypassPermissions
instead to allow full tool access in this controlled test environment.
</pre>
</div>
</content>
</entry>
</feed>
