Install
$ agentstack add skill-tamdogood-builder-essential-skills-code-standards ✓ 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
Code standards — the change workflow
You are making a code change. This skill is the execution order; it is stack-agnostic. Adapt every command to the project's actual toolchain: read the project's own docs (README, CONTRIBUTING, AGENTS.md/CLAUDE.md, a Makefile or package scripts) once to learn its build, test, lint, and format commands before you start.
Steps
- Orient. Read any project lessons/gotchas file (e.g.
tasks/lessons.md) and obey it.
Find and read the one doc for the subsystem you're changing, not the whole tree.
- Plan if non-trivial (3+ steps or an architectural choice): short plan in
tasks/todo.md—
files touched, tests to add, risks. Going sideways → stop and re-plan.
- Baseline: get the project's build + test command green before touching anything. A
pre-existing red is the first task; never build on top of a broken baseline.
- Implement the smallest root-cause change. Hand-match the surrounding style.
Don't run a mass auto-formatter on files you touch; it buries your real diff in noise.
- Test: new behavior gets a test that fails without the change; security gates get
negative-assertion tests (wrong token → denied, state unchanged, no data leak).
- Verify: run the full check suite (build + lint + type-check + tests) until green
before you call it "done".
- Self-review your own diff as if it were someone else's PR (or invoke a review
skill). Fix findings before presenting.
- Close out: update docs in the same change; append to a lessons file if you were
corrected or surprised; write a conventional commit message (feat|fix|docs(scope): …).
Tripwires — stop and re-check the project's guidelines when you're about to…
- change a public or shared interface (API, wire format, schema) → check every caller; prefer additive changes
- add a dependency → weigh transitive weight and licenses, and whether the stdlib already covers it
- touch auth, tokens, or permissions → audit every writer and reader path, add negative tests
- write a secret to disk → use the platform's atomic create-private API, never write-then-chmod
- hold a lock across an
await/blocking call, or read env inside logic → extract a pure decision function
No-progress guard
The same failure surviving two fix attempts means stop: write the finding to your lessons file, mark the item [BLOCKED], and surface it. Don't thrash.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: tamdogood
- Source: tamdogood/builder-essential-skills
- License: MIT
- Homepage: https://www.npmjs.com/package/@tamng0905/builder-essential-skills
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.