Install
$ agentstack add skill-steph-dove-klaussy-agents-fastapi-fix ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Fix all lint, format, and type errors in the current changes.
Steps
- Read CLAUDE.md to find the project's lint, format, and type-check commands. If they're missing, fall back to the stack defaults below.
- Read any
.claude/rules/*.mdwhosepaths:glob matches the files you're about to touch — they may carry style/typing rules the linter doesn't enforce. - Scope to this branch's change. Build the changed-file list — the union of:
git diff --name-only master...HEAD(work committed on this branch)git diff --name-only(unstaged) andgit diff --name-only --cached(staged)
Pass these paths to every command below so the tools judge only what this branch changed — never the whole repo. Pre-existing violations in untouched files are not yours to fix here. If the list is empty, there's nothing to fix; say so and stop. (If master...HEAD errors because the base branch isn't present locally, fall back to the uncommitted diff alone.)
- Run each command in this order, scoped to the changed files (apply each pass before running the next, so fixes from one don't fight the next):
- Format first (it normalizes whitespace and quoting that lint rules might complain about).
- Lint next (it picks up real style/safety issues on top of formatted code).
- Type-check last (type errors don't move under format/lint, but the line numbers will). If the type-checker needs whole-project context to resolve imports, run it normally but only fix errors that land in the changed files.
- Fix all reported issues. When format and lint disagree on a specific construct, lint wins (formatters auto-resolve; lint encodes intent).
- Re-run all three (still scoped to the changed files) to verify they're clean.
Stack defaults (if CLAUDE.md doesn't specify commands)
Append the changed files to each command (e.g. ruff check ) so it stays scoped to the branch's change.
- Python:
ruff format,ruff check,mypy(orpyright) - TypeScript / JavaScript:
prettier --write,eslint --fix,tsc --noEmit(tscchecks the whole project — read only the changed files' errors) - Go:
gofmt -w/goimports -w,go vet,staticcheck - Rust:
cargo fmt,cargo clippy --fix,cargo check(cargo works per-crate — limit fixes to the changed files)
If none of these are configured (no pyproject.toml/package.json/go.mod/Cargo.toml entry, no devDependency installed), report that the repo has no configured lint/format/type-check toolchain and stop — don't pick one and force it on the project.
Rules
- Do NOT change logic or behavior. Only fix style, formatting, and type issues.
- If a type error reveals a real bug (e.g. a function call missing a required argument), STOP. Report it as a bug for the debug skill to take, don't paper over it with a cast or
Any. - If lint flags an existing violation in unchanged code (not from your diff), leave it alone unless explicitly asked to fix repo-wide.
- Remove commented-out code in the lines you're touching (the pre-commit guard blocks it). Don't add narrating comments while fixing — keep only short "why" comments.
When NOT to use
- The user wants behavior changes, refactors, or bug fixes — those are different skills.
- The repo has no linter/formatter/type-checker — there's nothing to run.
- The errors are runtime errors or test failures, not lint/format/type errors — use debug instead.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: steph-dove
- Source: steph-dove/klaussy-agents
- License: MIT
- Homepage: https://klaussy.com
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.