Weekly Engineering Retro
garrytan/gstack
Builds a weekly engineering retrospective from git history: commit counts, per-person contributions, work patterns and code quality numbers over a chosen window.
End-of-plan improvement kata. An agent skill from robertbarclayy/conducty.
Install Conducty Improve
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install robertbarclayy/conducty conducty-improve --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/conducty-improve .claude/skills/conducty-improve && rm -rf skills-srcUse ~/.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/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .claude/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improveType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install robertbarclayy/conducty conducty-improve --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/conducty-improve .agents/skills/conducty-improve && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .agents/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install robertbarclayy/conducty conducty-improve --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/conducty-improve .cursor/skills/conducty-improve && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .cursor/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/robertbarclayy/conducty.git --path skills/conducty-improve--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install robertbarclayy/conducty conducty-improve --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/conducty-improve .gemini/skills/conducty-improve && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .gemini/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install robertbarclayy/conducty conducty-improveInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/conducty-improve .github/skills/conducty-improve && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .github/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add robertbarclayy/conducty --skill conducty-improve -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install robertbarclayy/conducty conducty-improve --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/robertbarclayy/conducty.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/conducty-improve .opencode/skills/conducty-improve && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "conducty-improve" agent skill from https://github.com/robertbarclayy/conducty/tree/main/skills/conducty-improve into .opencode/skills/conducty-improve/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "conducty-improve", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
conducty-improve- GitHub stars
- 176
- Token cost
- ~1.5k tokens
- SKILL.md length
- 635 words
- Files
- 1
- Skills in repo
- 20
- Repo updated
- First seen
- Licence
- MIT
At a glance
End-of-plan improvement kata. An agent skill from robertbarclayy/conducty.
- Says what did we learn
- SKILL.md covers The Improvement Kata, Recording Improvements, What Makes This Different from… and Integration, plus 1 more section
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
- Improve the process
What it does
Conducty Improve is an agent skill from robertbarclayy/conducty. End-of-plan improvement kata. Extracts lessons from the just-finished plan, identifies experiments, and shapes the next plan's approach. Use after [[conducty-review]] completes, or when the user says "what did we learn", "improve the process", "retrospective", "wrap up this plan".
Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Product & Project Management, covering Retrospectives. The repository describes itself as: Stop context-switching. Start batch-planning. Conducty replaces the fragmented cycle of plan → prompt → wait → get distracted → review → fix → repeat with advanced batch planning. The licence is MIT.
When your agent uses it
- Says what did we learn
- Improve the process
- Wrap up this plan
Example prompts
- “what did we learn”
- “improve the process”
- “retrospective”
- “/conducty-improve”
What it can do on your machine
Read from SKILL.md and the folder at commit 64aefd5. 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 (its code samples are markdown).
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
Conducty Improve loads about 1.5k tokens when it runs. Until then it costs about 75 tokens; SKILL.md has 635 words of instructions outside code blocks.
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 robertbarclayy/conducty at commit 64aefd5, republished under its MIT licence (© robertbarclayy). 635 words, ~1,481 tokens.
Download SKILL.mdSave it as
.claude/skills/conducty-improve/SKILL.md (or your agent's skills folder).- name
- conducty-improve
- description
- End-of-plan improvement kata. Extracts lessons from the just-finished plan, identifies experiments, and shapes the next plan's approach. Use after [[conducty-review]] completes, or when the user says "what did we learn", "improve the process", "retrospective", "wrap up this plan".
- aliases
- conducty-improve, improve
- tags
- conducty/skill, conducty/improve
Conducty Improve — The Learning Loop
The improvement kata transforms execution data into process improvements. Without this step, Conducty is an execution pipeline that repeats the same mistakes. With it, Conducty is a learning system that gets better with every plan.
This is the step that closes the feedback loop: Shape → Plan → Execute → Verify → Improve → Shape (next plan).
The Improvement Kata
Adapted from Toyota Kata (Mike Rother) — four questions applied to agentic work.
Question 1: What Was the Target Condition?
Read the just-finished plan note from the vault (Plans/Plan YYYY-MM-DD HHmm [Topic].md — see [[conducty-obsidian]]):
- What goals were set?
- What was the appetite?
- What was the expected pass rate and velocity?
- What improvement experiments from the prior plan were being tested?
Question 2: What Is the Current Condition?
Read the plan's review results (the ## End-of-Plan Summary and ## Checkpoint Notes):
- What actually happened? How many prompts completed vs. failed?
- What was the first-attempt pass rate?
- How did actual time compare to appetite?
- Did the improvement experiments from the prior plan show results?
Question 3: What Obstacles Did We Encounter?
Read recent entries from [[Failure Patterns]] (the accumulating note in the vault):
- What patterns caused failures?
- At which leverage level were most failures? (plan / prompt / code)
- Which prompt smells were most common?
- Were there systemic issues?
- What surprised us?
Categorize obstacles:
| Category | Example | Fix Level |
|---|---|---|
| Stale context | Agent used outdated architecture info | Refresh [[conducty-context]] |
| Prompt smell | Vague acceptance criteria led to wrong implementation | Improve prompt template |
| Design gap | Shaping missed a key constraint | Improve [[conducty-shape]] process |
| Calibration error | Low complexity prompt turned out to be Medium | Adjust complexity estimation |
| Template weakness | Bug fix template didn't include characterization step | Fix the template |
| Tool limitation | Agent model couldn't handle the task complexity | Adjust model selection |
Question 4: What Is Our Next Experiment?
Propose 1-3 specific, testable experiments for the next plan:
Each experiment should be:
- Specific: "Add characterization step to ALL refactor prompts" not "improve prompts"
- Testable: You can tell next plan whether it worked
- Small: Changes one thing at a time
- Targeted: Addresses a specific obstacle from Question 3
Examples:
- "Last plan, 3 prompts failed because of missing context. Next: include the module dependency graph in context for prompts touching shared modules."
- "Last plan's pass rate was 60%. The main smell was vague acceptance criteria. Next: every acceptance criterion must include a specific number or behavior, no qualitative criteria."
- "Last plan's refactor prompts all needed extra retries. Next: add mandatory characterization step to every refactor prompt before any changes begin."
Show full SKILL.md (220 more words)Show less
Recording Improvements
Write a new note to the vault: Improvements/Improvement YYYY-MM-DD HHmm.md (timestamp = when the kata is run, typically right after [[conducty-review]]).
---
type: improvement
date: YYYY-MM-DD
time: HHmm
tags: [conducty, conducty/improvement]
---
# Improvement YYYY-MM-DD HHmm
**Target condition**: {what we planned}
**Current condition**: {what happened — pass rate, velocity, appetite usage}
## Obstacles
- {obstacle 1} — leverage: {plan/prompt/code}
- {obstacle 2} — leverage: {plan/prompt/code}
## Experiments for Next Plan
1. {specific experiment} — testing whether {hypothesis}
2. {specific experiment} — testing whether {hypothesis}
## Prior Experiments Evaluated
- {prior experiment 1}: {worked / didn't work / inconclusive} — {evidence}
## Related
- Index: [[Improvements Index]]
- Reflects on: [[Plan YYYY-MM-DD HHmm Topic]]
- Patterns: [[Failure Patterns]]
- Trends: [[Metrics]]
- Prior: [[Improvement YYYY-MM-DD HHmm]]Then prepend the new note's wikilink to [[Improvements Index]].
What Makes This Different from Logging
Logging records what happened. Improvement changes what happens next.
The difference is in Question 4. If you finish the kata without specific experiments that will change the next plan, you've just logged — you haven't improved. Every kata must produce at least one concrete experiment.
The test: Can [[conducty-plan]] read the latest improvement note and produce a measurably different plan? If yes, the kata worked. If no, you just wrote a diary entry.
Integration
- Input from [[conducty-review]]: End-of-plan summary, failure patterns, velocity metrics, carry-forward items
- Input from [[conducty-checkpoint]]: Health metrics, hill chart positions, systemic issues
- Output to [[conducty-plan]]: The next plan reads the latest
Improvement YYYY-MM-DD HHmmnote in Step 1 and applies the experiments - Output to prompt templates: If a template weakness is identified, edit the template file directly (Edit tool)
- Output to [[conducty-context]]: If stale context was an obstacle, trigger a context refresh
Frequency
- Per plan (primary): After [[conducty-review]], before starting the next plan
- Weekly (optional): Skim the week of improvement notes via [[Improvements Index]] for larger patterns
- After major failures: When a goal is blocked or pass rate drops below 50%
© robertbarclayy, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/conducty-improve of robertbarclayy/conducty.
Open the folder on GitHubat commit 64aefd5
Compare with similar skills
Conducty Improve 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Conducty Improve this skillrobertbarclayy/conducty | 176 | — | ~1.5k | Automated safety check: Pass | MIT | |
| Weekly Engineering Retrogarrytan/gstack | 136k | — | ~2.4k | Automated safety check: Pass | MIT | |
| Dough Execute Planterryyin/lizard | 2.5k | — | ~4.3k | Automated safety check: Pass | Custom licence | |
| Oral Paper SkillAdkid-Zephyr/oral-paper-skill | 357 | — | ~1.9k | Automated safety check: Pass | None | |
| Deck Retroasheshgoplani/agent-deck | 1.1k | — | ~1.8k | Automated safety check: Pass | MIT | |
| Dough Execution Retrospectiveterryyin/lizard | 2.5k | — | ~4k | Automated safety check: Pass | Custom licence |
Similar skills
Dough Execute Plan
terryyin/lizard
Executes one selected story or bounded retrospective correction through an executable plan, or one authorized planless slice from a selected simple story or a contextual instruction, with…
Oral Paper Skill
Adkid-Zephyr/oral-paper-skill
Help authors learn from exemplary ICLR, ICML, and NeurIPS papers through source-linked manuscript comparisons, concrete writing and experiment suggestions, and guided reflection.
Deck Retro
asheshgoplani/agent-deck
Run a fully local agent-deck retrospective over the user's own transcripts, Recall index and logs.
Dough Execution Retrospective
terryyin/lizard
Reviews planned, completed planless quick, or quick-to-planned execution against original intent, aggregate commits, current whole-product architecture, and tests, including after cleanup.
Criticism Self Criticism
HughYau/qiushi-skill
批评与自我批评：在工作完成、阶段验收、收到批评或同类错误反复出现时，对成果和过程做诚实、具体、基于事实的审视，输出可执行的改进项，并处理外来批评而不辩解。触发信号包括 review、复盘、审查、"帮我看看有没有问题"、"你确定吗"；任务刚开始或只是单步查询时不触发。
More from robertbarclayy/conducty
All 20 skills in this repoConducty Checkpoint
robertbarclayy/conducty
Quality gate between parallelization groups. An agent skill from robertbarclayy/conducty.
Conducty Code Review
robertbarclayy/conducty
Standalone post-implementation code review across a whole branch or PR.
Conducty Context
robertbarclayy/conducty
Ingest a project directory into the Obsidian vault as a linked context sub-graph — Architecture, Conventions, Invariants, Hotspots, Tests, Glossary, plus per-bounded-context module notes.
Conducty Debug
robertbarclayy/conducty
Leverage-point analysis for failed prompts. An agent skill from robertbarclayy/conducty.
Conducty Execute
robertbarclayy/conducty
Dispatches Claude Code Task subagents to run prompts from a vault plan note.
Conducty Plan
robertbarclayy/conducty
Batch planning of AI prompts. An agent skill from robertbarclayy/conducty.
Related
Categories
Questions about Conducty Improve
What does Conducty Improve do?
End-of-plan improvement kata. An agent skill from robertbarclayy/conducty. Conducty Improve is an agent skill from robertbarclayy/conducty. End-of-plan improvement kata.
When should I use Conducty Improve?
Conducty Improve fits situations like: says what did we learn; improve the process; wrap up this plan.
How do I install Conducty Improve in Claude Code?
Run `npx skills add robertbarclayy/conducty --skill conducty-improve -a claude-code`. Or copy the skill folder (skills/conducty-improve in robertbarclayy/conducty) into .claude/skills/conducty-improve in your project. Claude Code loads it when a task matches its description.
How do I install Conducty Improve in Codex?
Run `npx skills add robertbarclayy/conducty --skill conducty-improve -a codex`. Or copy the skill folder (skills/conducty-improve in robertbarclayy/conducty) into .agents/skills/conducty-improve in your project. Codex loads it when a task matches its description.
Can I use Conducty Improve 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 robertbarclayy/conducty --skill conducty-improve -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/conducty-improve, .gemini/skills/conducty-improve, .github/skills/conducty-improve and .opencode/skills/conducty-improve in your project.
What does Conducty Improve need to run?
SKILL.md names no scripts, command-line tools or credentials: Conducty Improve is instructions for the agent only.
Does Conducty Improve 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 Conducty Improve 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 Conducty Improve use?
Conducty Improve is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Conducty Improve use?
About 1.5k tokens (SKILL.md is roughly 5.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Conducty Improve?
Skills that share tags, products or a category with Conducty Improve: Weekly Engineering Retro (garrytan/gstack, 136k stars), Dough Execute Plan (terryyin/lizard, 2.5k stars), Oral Paper Skill (Adkid-Zephyr/oral-paper-skill, 357 stars) and Deck Retro (asheshgoplani/agent-deck, 1.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Conducty Improve?
robertbarclayy (a GitHub user) maintains it in robertbarclayy/conducty, which has 176 GitHub stars. The repository holds 20 skills in this directory. The repository was last updated on June 19, 2026.
Source: robertbarclayy/conducty on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.