一键在 Manus 中运行任何 Skill

fix-issue

星标11

分支2

更新时间2026年5月17日 05:46

Investigate and fix a GitHub issue in this repo — pull the issue with `gh`, trace it to the root cause across `lib/` and `bin/`, fix portably, and cover it with a bats regression test. Use this skill when the user asks to fix, resolve, or tackle a GitHub issue (by number or URL), debug a reported bug from the tracker, or work through an issue end-to-end — including phrasings like "fix #42", "tackle issue 17", "resolve this GH issue", "почини issue #N", or "разберись с багом из тикета".

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

yelmuratoff

yelmuratoff/agent_sync

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

Fix Issue

Resolve a reported GitHub issue: understand it, trace to the root cause, fix portably, and lock the fix in with a bats test.

Steps

Pull the issue — gh issue view <number> for title, body, comments. If the user pasted a URL, extract the number from it.
Reproduce — run the failing flow against a fresh fixture (use setup_test_project from tests/test_helper.bash so you don't contaminate the working tree). A bug you can't reproduce is a bug you can't verify fixed.
Trace to the root cause — work outward from the symptom: bin/agentsync.sh (command routing) → lib/sync.sh (per-tool loop) → lib/helpers/*.sh (file ops, YAML parsing, rule merging, format conversion, path safety) → .ai/src/tools/*.yaml. Don't stop at the first plausible cause.
Fix with the smallest correct change — touch only the code that owns the bug. Three similar lines beat a premature abstraction.
Keep it portable — POSIX sed (write-then-mv), cd "$(dirname "$path")" && pwd instead of realpath, quoted "$var", local in functions, set -euo pipefail stays on.
Lint — shellcheck -x -S warning -e SC1091 <changed-scripts>. Resolve warnings; don't paper over with # shellcheck disable=.
Add a bats regression test — fails on main, passes on the fix. Name by behaviour: @test "sync: <what the bug broke> stays correct". Hermetic temp dirs via setup_test_project / teardown_test_project.
Run the full suite — bats tests/. Every test green, not just the new one.
Commit referencing the issue — fix: <subject> (#<issue-number>) so GitHub auto-closes on merge.

Gotchas

Issue authors describe the symptom, not the cause. "Sync deletes my file" might actually be "cleanup runs on a stale manifest" — fix the chain, not the wording.
The suite runs on macOS, Linux, and Git Bash on Windows in CI. A test passing on your macOS shell with sed -i '' will fail on Linux.
Tool-specific bugs (cursor, codex, …) — pin the regression test to that tool, but run the full suite so the fix doesn't regress the others.
Idempotency bugs (sync output differs on the second run) are usually find | sort missing LC_ALL=C, or temp filenames based on PID/timestamp. Look there first.
A fix without a test is a regression waiting to happen. Skipping the test step is non-negotiable.

同仓库更多 Skills

同仓库

agentsync

yelmuratoff/agent_sync

Create or edit AgentSync configuration — AGENTS.md, rules, skills, commands, subagents, settings, MCP servers, hooks, or per-tool configs. Use this skill when adding a rule, creating or scaffolding a skill, writing a slash command, defining a subagent persona, editing permissions, configuring an MCP server, setting up the `.ai/src/` directory, or running `agentsync sync` / `add` / `customize` / `resolve` / `simplify` / `profile` — even when the user does not name "AgentSync" explicitly but is editing files in `.ai/src/`, `.claude/`, `.cursor/`, or another tool-config directory.

2026-07-2011

agentsync

yelmuratoff/agent_sync

2026-07-0411

release

yelmuratoff/agent_sync

Bump the AgentSync version, write a CHANGELOG entry summarising changes since the last tag, and prepare a clean release commit. Use this skill when the user asks to release, ship, cut a version, bump major/minor/patch, tag, or prepare a new version — including phrasings like "let's ship 0.12", "поднять версию", "сделай релиз", or when asked to update CHANGELOG.md after a stretch of work.

2026-06-3011

content

yelmuratoff/agent_sync

One imperative sentence on what this skill does + concrete trigger conditions. Be pushy — list contexts including ones where the user doesn't name the domain ("even when phrased as 'X' or 'Y'"). Pack relevant keywords. Stay under 1024 characters.

2026-06-1311

commit

yelmuratoff/agent_sync

Create a well-structured git commit for staged or unstaged changes — analyze the diff, match the project's existing commit style, write a clear message, and create the commit. Use this skill when the user asks to commit, save changes, check in code, write a commit message, prepare changes for push, or wrap up a piece of work — even when they don't say "commit" explicitly (e.g. "let's save this", "ship it", "сохрани изменения").

2026-06-0311

prompt-engineering

yelmuratoff/agent_sync

Design, debug, and refine prompts for LLM-based coding tools, agents, and pipelines. Use this skill when writing or rewriting a system prompt, agent persona, slash-command body, rule, or skill description; tuning a prompt that produces vague, incomplete, off-format, hallucinated, or over-/under-triggering output; choosing among techniques like XML tags, few-shot examples, role-setting, reasoning steps, or recap; debugging a model that ignores instructions, leaks information, or over-engineers solutions; or designing the persona, autonomy posture, or tool-use rules of an autonomous agent — even when the user does not explicitly mention "prompt engineering" or "system prompt". Not for AgentSync file structure, frontmatter, or scaffolding new rule/skill/command files — use the `agentsync` skill for that.

2026-05-3011

name

fix-issue

description

Fix Issue

Resolve a reported GitHub issue: understand it, trace to the root cause, fix portably, and lock the fix in with a bats test.

Steps

Pull the issue — gh issue view <number> for title, body, comments. If the user pasted a URL, extract the number from it.
Reproduce — run the failing flow against a fresh fixture (use setup_test_project from tests/test_helper.bash so you don't contaminate the working tree). A bug you can't reproduce is a bug you can't verify fixed.
Trace to the root cause — work outward from the symptom: bin/agentsync.sh (command routing) → lib/sync.sh (per-tool loop) → lib/helpers/*.sh (file ops, YAML parsing, rule merging, format conversion, path safety) → .ai/src/tools/*.yaml. Don't stop at the first plausible cause.
Fix with the smallest correct change — touch only the code that owns the bug. Three similar lines beat a premature abstraction.
Keep it portable — POSIX sed (write-then-mv), cd "$(dirname "$path")" && pwd instead of realpath, quoted "$var", local in functions, set -euo pipefail stays on.
Lint — shellcheck -x -S warning -e SC1091 <changed-scripts>. Resolve warnings; don't paper over with # shellcheck disable=.
Add a bats regression test — fails on main, passes on the fix. Name by behaviour: @test "sync: <what the bug broke> stays correct". Hermetic temp dirs via setup_test_project / teardown_test_project.
Run the full suite — bats tests/. Every test green, not just the new one.
Commit referencing the issue — fix: <subject> (#<issue-number>) so GitHub auto-closes on merge.

Gotchas

Issue authors describe the symptom, not the cause. "Sync deletes my file" might actually be "cleanup runs on a stale manifest" — fix the chain, not the wording.
The suite runs on macOS, Linux, and Git Bash on Windows in CI. A test passing on your macOS shell with sed -i '' will fail on Linux.
Tool-specific bugs (cursor, codex, …) — pin the regression test to that tool, but run the full suite so the fix doesn't regress the others.
Idempotency bugs (sync output differs on the second run) are usually find | sort missing LC_ALL=C, or temp filenames based on PID/timestamp. Look there first.
A fix without a test is a regression waiting to happen. Skipping the test step is non-negotiable.