<feed xmlns='http://www.w3.org/2005/Atom'>
<title>superpowers.git, branch codex-sync-excludes-and-porting-examples</title>
<subtitle>Agentic skills framework and development workflow
</subtitle>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/'/>
<entry>
<title>docs(porting): Appendix A shows examples, not every harness</title>
<updated>2026-10-08T02:27:14+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2026-10-08T02:27:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=e4dc1ced68b7eef84f61f040502fdfb03db8fc50'/>
<id>e4dc1ced68b7eef84f61f040502fdfb03db8fc50</id>
<content type='text'>
It called itself the live index of reference integrations, which it can't
stay without an update for every new harness, and it had already fallen
behind (no Devin, Hermes, or Muse). Present it as worked examples of each
integration shape and point to the repo's manifests as the source of truth.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It called itself the live index of reference integrations, which it can't
stay without an update for every new harness, and it had already fallen
behind (no Devin, Hermes, or Muse). Present it as worked examples of each
integration shape and point to the repo's manifests as the source of truth.
</pre>
</div>
</content>
</entry>
<entry>
<title>fix(codex-sync): exclude the Hermes and Muse plugin manifests</title>
<updated>2026-10-08T02:27:00+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@fsck.com</email>
</author>
<published>2026-10-08T02:27:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=eabbe888edf50ebb39d5bf7e02eb1aa860a864b1'/>
<id>eabbe888edf50ebb39d5bf7e02eb1aa860a864b1</id>
<content type='text'>
The Codex sync excluded every other harness's dotdir but shipped
.hermes-plugin/ and .muse-plugin/ into the Codex plugin.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The Codex sync excluded every other harness's dotdir but shipped
.hermes-plugin/ and .muse-plugin/ into the Codex plugin.
</pre>
</div>
</content>
</entry>
<entry>
<title>brainstorming: rebuild around drawing out intent (#2463)</title>
<updated>2026-10-06T23:57:52+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-10-06T23:57:52+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=9e639188a1a8029926bda6e5d7edbf66ef3ac333'/>
<id>9e639188a1a8029926bda6e5d7edbf66ef3ac333</id>
<content type='text'>
* docs: intent doc for brainstorming rebuild

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* docs: spike exemption to the brainstorming gate

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: rebuild around drawing out intent

Replaces the spike/bounded/architectural router with an intent-first
conversation: open describing questions, recon offers, playback in
200-300 word chunks, then sizing (quick task / small change / project
with a written design). The spec reviewer prompt becomes a builder
check that asks what a builder would still need to know.

Adds tests/brainstorming/first-turn.sh, a first-turn micro-test with
no-skill and skill arms.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: multi-turn brainstorming driver with simulated humans

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: harder small-change scenario; clearer launch failure

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: tighter opening, wait for builder check, reason from the why

Round 1 multi-turn runs: the builder check went to a background subagent
and stalled the conversation; the opening restated the request; one run
proposed CLI flags after hearing the tool runs from cron.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: builder check gates the handoff, not the reply; README

Claude Code subagents run in the background, so the subject can't block on
the builder check. The skill now says what matters (don't hand over the
document until the check is back), and converse.sh lets the simulated
human wait for the subject's follow-up turn instead of nudging it.
Also fixes converse.sh dying before its summary when the work dir has no
commits.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: split multi-project requests and take them one at a time

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: show it when the playback describes something visual

Across 13 multi-turn runs neither skill ever offered the visual companion,
even for a kids' tablet screen. Tying the offer to an observable moment
(talk turns to how something looks, or a playback chunk describes it) got
offers in 2/3 runs, each as its own message, none upfront. The simulated
human now declines browser offers so test runs don't open tabs.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: opt-out, spike, and multi-project brainstorming scenarios

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: resolve review findings from adversarial pass

- Opt-out ends the skill and the gate with it (the gate had no rule for it).
- Each message ends with one question, no example answers hanging off it;
  multi-question messages and parenthetical menus showed up in most runs.
- Playback marks the agent's guesses as guesses; one run played the
  agent's own example back as the human's.
- Restore the old gate's list of what counts as building (#2258).
- Companion offer stays its own message; mockups only after acceptance.
- converse.sh: survive a failed human turn, report why a run ended, and
  keep raw task notifications out of the simulated human's view.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: play back as you go, alternating with questions

With playback saved for the end, project brainstorms ran 8-15 questions
before the human saw any summary. Interleaved, the first playback lands
around turn 3-6 (6 runs: habit-tracker, pottery, offsite x2), so
corrections arrive before more questions build on a wrong guess. Total
length is about the same.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: blind-pair launcher for comparing brainstorming versions

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: builder check yields one ranked round of questions

In a blind comparison (6 testers, old vs new skill), every new-skill tester
flagged the builder check: two rounds and 3-9 questions after they'd said
'write it up'. Now the agent warns once that a reviewer reads first,
settles minor gaps itself and marks them as its calls, and brings only the
questions that change what gets built, in one message, most important
first. 4/4 runs (habit-tracker, pottery x2) showed a single round.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* docs: brainstorming rebuild eval results

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* Move the eval harness and results to their own PR

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: one line on writing style

Blind testers flagged a researched answer as a wall of text for a
non-technical person. Micro-tested 12 style lines on research reports and
mid-conversation answers (3 prompts x 5 reps each). Every variant already
led with the point; naming a journalist (wire reporter, Mossberg, a
Journalism BA) added source attributions. Plain wording shortened research
reports by about 12% against no line.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: offer the visual companion whenever seeing would help

Lists the moments that call for it (things they'll look at, layouts,
flows, options that look different, diagrams/timelines/maps), checks each
playback chunk before sending it, and drops the 'token-intensive' line
that confused non-technical blind testers. Offers: habit tracker 3/3 (was
~50%), pottery booking 3/3 (was 0/3), at turns 3-7, never upfront; 0/4 on
scenarios with nothing visual (sync-tool flag, talk outline).

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: plain style line

Tested five style lines across Opus, Sonnet, gpt-6-luna and gpt-6-astra
(3 prompts x 5 reps each); none moved length or attribution beyond noise
on any model. Keep the plainest statement of intent.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: report the most important findings, offer the rest

Research reports to a non-technical person ran 320-400 words on Claude.
With this line, replies that offer the rest rose from 1/5 to 4/5 (Opus)
and 3/5 (Sonnet); Opus reports ~7% shorter; gpt-6-luna/astra, already
~170-190 words, unchanged.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
* docs: intent doc for brainstorming rebuild

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* docs: spike exemption to the brainstorming gate

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: rebuild around drawing out intent

Replaces the spike/bounded/architectural router with an intent-first
conversation: open describing questions, recon offers, playback in
200-300 word chunks, then sizing (quick task / small change / project
with a written design). The spec reviewer prompt becomes a builder
check that asks what a builder would still need to know.

Adds tests/brainstorming/first-turn.sh, a first-turn micro-test with
no-skill and skill arms.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: multi-turn brainstorming driver with simulated humans

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: harder small-change scenario; clearer launch failure

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: tighter opening, wait for builder check, reason from the why

Round 1 multi-turn runs: the builder check went to a background subagent
and stalled the conversation; the opening restated the request; one run
proposed CLI flags after hearing the tool runs from cron.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: builder check gates the handoff, not the reply; README

Claude Code subagents run in the background, so the subject can't block on
the builder check. The skill now says what matters (don't hand over the
document until the check is back), and converse.sh lets the simulated
human wait for the subject's follow-up turn instead of nudging it.
Also fixes converse.sh dying before its summary when the work dir has no
commits.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: split multi-project requests and take them one at a time

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: show it when the playback describes something visual

Across 13 multi-turn runs neither skill ever offered the visual companion,
even for a kids' tablet screen. Tying the offer to an observable moment
(talk turns to how something looks, or a playback chunk describes it) got
offers in 2/3 runs, each as its own message, none upfront. The simulated
human now declines browser offers so test runs don't open tabs.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: opt-out, spike, and multi-project brainstorming scenarios

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: resolve review findings from adversarial pass

- Opt-out ends the skill and the gate with it (the gate had no rule for it).
- Each message ends with one question, no example answers hanging off it;
  multi-question messages and parenthetical menus showed up in most runs.
- Playback marks the agent's guesses as guesses; one run played the
  agent's own example back as the human's.
- Restore the old gate's list of what counts as building (#2258).
- Companion offer stays its own message; mockups only after acceptance.
- converse.sh: survive a failed human turn, report why a run ended, and
  keep raw task notifications out of the simulated human's view.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: play back as you go, alternating with questions

With playback saved for the end, project brainstorms ran 8-15 questions
before the human saw any summary. Interleaved, the first playback lands
around turn 3-6 (6 runs: habit-tracker, pottery, offsite x2), so
corrections arrive before more questions build on a wrong guess. Total
length is about the same.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* tests: blind-pair launcher for comparing brainstorming versions

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: builder check yields one ranked round of questions

In a blind comparison (6 testers, old vs new skill), every new-skill tester
flagged the builder check: two rounds and 3-9 questions after they'd said
'write it up'. Now the agent warns once that a reviewer reads first,
settles minor gaps itself and marks them as its calls, and brings only the
questions that change what gets built, in one message, most important
first. 4/4 runs (habit-tracker, pottery x2) showed a single round.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* docs: brainstorming rebuild eval results

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* Move the eval harness and results to their own PR

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: one line on writing style

Blind testers flagged a researched answer as a wall of text for a
non-technical person. Micro-tested 12 style lines on research reports and
mid-conversation answers (3 prompts x 5 reps each). Every variant already
led with the point; naming a journalist (wire reporter, Mossberg, a
Journalism BA) added source attributions. Plain wording shortened research
reports by about 12% against no line.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: offer the visual companion whenever seeing would help

Lists the moments that call for it (things they'll look at, layouts,
flows, options that look different, diagrams/timelines/maps), checks each
playback chunk before sending it, and drops the 'token-intensive' line
that confused non-technical blind testers. Offers: habit tracker 3/3 (was
~50%), pottery booking 3/3 (was 0/3), at turns 3-7, never upfront; 0/4 on
scenarios with nothing visual (sync-tool flag, talk outline).

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: plain style line

Tested five style lines across Opus, Sonnet, gpt-6-luna and gpt-6-astra
(3 prompts x 5 reps each); none moved length or attribution beyond noise
on any model. Keep the plainest statement of intent.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK

* brainstorming: report the most important findings, offer the rest

Research reports to a non-technical person ran 320-400 words on Claude.
With this line, replies that offer the rest rose from 1/5 to 4/5 (Opus)
and 3/5 (Sonnet); Opus reports ~7% shorter; gpt-6-luna/astra, already
~170-190 words, unchanged.

Claude-Session: https://claude.ai/code/session_01M6AdHKvzoBUPTxPnYDwssK</pre>
</div>
</content>
</entry>
<entry>
<title>docs(release-notes): Hermes skill descriptions (#2284)</title>
<updated>2026-09-27T02:37:45+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:37:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=b1f8774768021fc389ccbde678c4a56e91bd76e6'/>
<id>b1f8774768021fc389ccbde678c4a56e91bd76e6</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>fix(hermes): pass each skill's frontmatter description to register_skill</title>
<updated>2026-09-27T02:37:44+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:37:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=923b30043562d0767bf06006f15224d1861adb63'/>
<id>923b30043562d0767bf06006f15224d1861adb63</id>
<content type='text'>
Hermes lists plugin skills (skills_list) with the description given to
register_skill, which defaults to "". The plugin registered only a
name and path, so all fourteen superpowers skills appeared with blank
descriptions and the model had nothing to choose on. Read each
SKILL.md's description and pass it through.

From #2284 by @pepijn-blom.

Co-authored-by: pepijn-blom &lt;322216308+pepijn-blom@users.noreply.github.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Hermes lists plugin skills (skills_list) with the description given to
register_skill, which defaults to "". The plugin registered only a
name and path, so all fourteen superpowers skills appeared with blank
descriptions and the model had nothing to choose on. Read each
SKILL.md's description and pass it through.

From #2284 by @pepijn-blom.

Co-authored-by: pepijn-blom &lt;322216308+pepijn-blom@users.noreply.github.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>docs(release-notes): plan markers stay repo-relative on Windows</title>
<updated>2026-09-27T02:31:12+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:31:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=874fd4255d50202055aac8d20a20d6c23af94d85'/>
<id>874fd4255d50202055aac8d20a20d6c23af94d85</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>test(executing-plans): compare the brief path in the scripts' own spelling on Git Bash</title>
<updated>2026-09-27T02:30:38+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:30:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=49b1db53d86d358a0a60446ccecba81c7595f7d0'/>
<id>49b1db53d86d358a0a60446ccecba81c7595f7d0</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>sdd-workspace: resolve the repo root physically so plan markers stay repo-relative on Git Bash</title>
<updated>2026-09-27T02:29:40+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:29:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=2b7893a1993a386755fb0907fa1a849112ca0a6c'/>
<id>2b7893a1993a386755fb0907fa1a849112ca0a6c</id>
<content type='text'>
git rev-parse --show-toplevel prints C:/Users/... on Windows while the
plan path comes from pwd -P (/c/Users/... or /tmp/...), so the
repo-relative check never matched and every marker held an absolute,
machine-specific path. Found while landing #2149's test fix on Windows.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
git rev-parse --show-toplevel prints C:/Users/... on Windows while the
plan path comes from pwd -P (/c/Users/... or /tmp/...), so the
repo-relative check never matched and every marker held an absolute,
machine-specific path. Found while landing #2149's test fix on Windows.
</pre>
</div>
</content>
</entry>
<entry>
<title>test(sdd-workspace): compare physical paths so the suite passes on Git Bash</title>
<updated>2026-09-27T02:28:31+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:28:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=c88ba76bdbdb3ebe5c29401854e9129509b2866d'/>
<id>c88ba76bdbdb3ebe5c29401854e9129509b2866d</id>
<content type='text'>
On Windows, git rev-parse --show-toplevel prints C:/Users/... while
the script (and pwd) print /tmp/..., so nine assertions compared two
spellings of the same directory and failed. The script itself was
correct. Normalize the expected paths the same way the script's output
is produced.

From #2149 by @t0domanh.

Co-authored-by: t0domanh &lt;3890425+t0domanh@users.noreply.github.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
On Windows, git rev-parse --show-toplevel prints C:/Users/... while
the script (and pwd) print /tmp/..., so nine assertions compared two
spellings of the same directory and failed. The script itself was
correct. Normalize the expected paths the same way the script's output
is produced.

From #2149 by @t0domanh.

Co-authored-by: t0domanh &lt;3890425+t0domanh@users.noreply.github.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>test(brainstorm-server): run shell helpers without a login shell</title>
<updated>2026-09-27T02:28:19+00:00</updated>
<author>
<name>Jesse Vincent</name>
<email>jesse@primeradiant.com</email>
</author>
<published>2026-09-27T02:28:19+00:00</published>
<link rel='alternate' type='text/html' href='https://git.1991861.xyz/superpowers.git/commit/?id=a0415daa1f33b00f3fe300f5c250a5beefafce4d'/>
<id>a0415daa1f33b00f3fe300f5c250a5beefafce4d</id>
<content type='text'>
The lifecycle tests ran their helpers with `bash -lc`, so anything a
login profile prints landed in captured output (a path from mktemp,
the newest session dir) and broke the test. Plain `bash -c` keeps the
helpers independent of the user's shell setup. Fixes #2190.

From #2192 by @tugrulguner.

Co-authored-by: tugrulguner &lt;47668043+tugrulguner@users.noreply.github.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The lifecycle tests ran their helpers with `bash -lc`, so anything a
login profile prints landed in captured output (a path from mktemp,
the newest session dir) and broke the test. Plain `bash -c` keeps the
helpers independent of the user's shell setup. Fixes #2190.

From #2192 by @tugrulguner.

Co-authored-by: tugrulguner &lt;47668043+tugrulguner@users.noreply.github.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
