fix-problem
Remediation protocol for resolving obstacles one at a time via hypothesize, test, fix, and document.
Skill by wonderbird
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/wonderbird-fix-problem
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, testing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/fix-problem
What's inside
- Skill name
fix-problem- Description
- Remediation protocol for resolving obstacles one at a time via hypothesize, test, fix, and document. Use when an unexpected obstacle prevents progress.
- License
- GPL-3.0
- Instructions
- 60 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Remediation Protocol
This protocol must be activated immediately upon discovering any unexpected
obstacle that prevents progress.
## Examples of Obstacles
- An unexpected test failure.
- A regression in functionality.
- A drop in code quality metrics (e.g., surviving mutants).
- An error or incorrect behavior from a command-line tool.
## Principle: One Obstacle at a Time
All work must be reduced to resolving a single, isolated obstacle. Attempting to
solve multiple issues simultaneously is strictly forbidden.
## Workflow
1. **STOP and Document:** Immediately stop the current primary task. Create a
new task representing the problem, blocking the original task. Claim the new
task and move it to in progress.
2. **Isolate and Prioritize:**
- List all identified obstacles in the new task
- Select **only the first obstacle** from the list. The remaining obstacles
must be ignored until the first is fully resolved.
3. **Execute a Single Fix Cycle:**
- **Hypothesize:** Formulate a specific, testable hypothesis for the cause
of the single obstacle.
- **Test Hypothesis (Red/Verification):** Create the minimal change
required to test the hypothesis.
- For a **code bug**, this is a minimal failing test (the "Red" step).
- For a **tooling bug**, this is a minimal command execution to see if
the new syntax works.
- **Verify Outcome:** Run the test or command.
- If it behaves as expected (e.g., the test fails, the command succeeds),
the hypothesis is likely correct. Proceed to the next step.
- If it does not behave as expected, the hypothesis is wrong. Revert the
change and return to the "Hypothesize" step to formulate a new one.
- **Implement Fix (Green):**
- For a **code bug**, write the minimal production code to make the test
pass.
- For a **tooling bug**, the fix was already verified in the previous
step. No further action is needed.
- **Verify Fix:** Run the relevant tests or commands to confirm the obstacle
is resolved and no regressions were introduced.
4. **Commit and Document:** Use the commit skill in a sub-agent to commit the
change with a message that clearly states which specific obstacle was
resolved. Update the task description to reflect the resolved obstacle and
any remaining ones.
5. **Repeat or Resume:**
- If there are more obstacles in the list from step 2, return to step 2 and
select the next one.
- If all obstacles are resolved, announce that the remediation is complete.
Close the remediation task and resume the original, primary task.From wonderbird/ai-agent-workspace, GPL-3.0.
More from wonderbird/ai-agent-workspace
- 0.598score
record-findings
wonderbird
Use when you want to keep track of important follow-up tasks and findings that arise during the software development and review process.
SkillAFreshness: A★ 2Claude Code +9 - 0.597score
kata-from-github-issue
wonderbird
Use when a new coding exercise - a kata - shall be crafted from a GitHub issue.
SkillAFreshness: A★ 2Claude Code +9 - 0.589score
commit
wonderbird
Prepares and executes a git commit
SkillAFreshness: A★ 2Claude Code +9 - 0.563score
conductor
wonderbird
Run a working session as a conductor who delegates every task to subagents and never implements anything personally.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
memory-update
wonderbird
Update memory bank files to keep them accurate for the active development.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
my-omc-parallel-work
wonderbird
Spawn parallel sub-agents in isolated git worktrees for independent Beads tasks.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
my-omc-show-next
wonderbird
Triage open issues down to the single next action.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
format-docs
wonderbird
Apply document style rules to markdown files such as meeting minutes, confluence pages, and project documentation.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
memory-create
wonderbird
Creates a focused memory bank for the current project iteration using scope, git history, and tests.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
memory-to-docs
wonderbird
Use when learnings in the memory bank should be preserved beyond the current development focus by moving them to more permanent documentation.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
memory-bank-by-cline
wonderbird
Memory bank concept and structure for maintaining project knowledge across sessions.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
code-review
wonderbird
Check code quality according to best practices and standards.
SkillBFreshness: B★ 2Claude Code +9
Signals
- Stars
- 2
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 7, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- GPL-3.0
- Repository
- github.com/wonderbird/ai-agent-workspace
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from wonderbird
- 0.598score
record-findings
wonderbird
Use when you want to keep track of important follow-up tasks and findings that arise during the software development and review process.
SkillAFreshness: A★ 2Claude Code +9 - 0.597score
kata-from-github-issue
wonderbird
Use when a new coding exercise - a kata - shall be crafted from a GitHub issue.
SkillAFreshness: A★ 2Claude Code +9 - 0.589score
commit
wonderbird
Prepares and executes a git commit
SkillAFreshness: A★ 2Claude Code +9 - 0.563score
conductor
wonderbird
Run a working session as a conductor who delegates every task to subagents and never implements anything personally.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
memory-create
wonderbird
Creates a focused memory bank for the current project iteration using scope, git history, and tests.
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
format-docs
wonderbird
Apply document style rules to markdown files such as meeting minutes, confluence pages, and project documentation.
SkillBFreshness: B★ 2Claude Code +9
Cite this page
fix-problem skill. Agent Mart. https://agentmart.dev/skills/wonderbird-fix-problem (last updated 2026-10-24)