Skip to content
Agentmart
Submit a tool

code-audit

Self-review implementation and affected tests in the current author context; never launch or act as an independent auditor.

Skill by tigger-developer

IndexedApache-2.0
Write a review

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/tigger-developer-code-audit

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.

We use your email only for this newsletter. Privacy

About

Self-review implementation and affected tests in the current author context; never launch or act as an independent auditor.

Use cases: security

No 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/code-audit

What's inside

Skill name
code-audit
Description
Self-review implementation and affected tests in the current author context; never launch or act as an independent auditor.
License
Apache-2.0
Instructions
44 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

Use only for a workflow-required or explicitly requested self-audit. Review in
the author's current context; do not delegate or select an independent role.
Read ~/.agents/sdlc/SELF-AUDIT.md and REVIEW.md's delivery-code criteria, with
READING's retained-context reuse. Inspect original candidate evidence rather
than only the authoring summary. This skill never loads the independent-auditor
entry point or launches the retired audit harness.

Select the mode explicitly:

- **SDLC ticket:** assess the approved requirements and design. Before review,
  required checks must have finished and passed, required AT/RT/OT records must be
  GREEN and the installed validator's delivery-code readiness check must pass.
  UTs may remain AMBER for the operator. Review affected product documentation.
- **Paired/emergency:** use recorded operator-authorized scope without imposing
  an advance definition gate. Required product documentation is current; the
  retrospective specification may follow under the governing workflow.
- **Spike:** use the brief and current exploratory intent after implementation
  is written, accepting the caller's timing declaration. All coding and testing
  standards apply, without full delivery gates/documentation requirements. No
  definition audit, merge or deployment is authorized by this review.
- **Non-SDLC:** assess supplied project context and coding/testing standards;
  bypass automatic SDLC readiness and spec/work/validation artefact requirements.
  Do not substitute invented requirements for missing context.

Review implemented tests first, then production code. Check correctness,
security, regressions, real invocation boundaries, meaningful assertions,
coverage, dependency risks, maintainability and approved design conformance,
including REVIEW's criteria for recorded hands-off departures.
Check state ownership, failure handling, concurrency and compatibility where
relevant. Passing tests do not prove design conformance or adequate coverage.
Do not use source-grep tests as behavioural evidence.

Identify unauthorized scope or unnecessary complexity in the delta without
reopening approved design choices. A blocking finding needs a concrete defect
or violated requirement/decision/standard. Optional refactoring and stylistic
preferences are suggestions, not reasons to fail. Report insufficient evidence
and material unresolved decisions honestly.

Report every blocking finding together with location, evidence and correction.
Remediate within scope, rerun affected checks and reassess under SELF-AUDIT.
Record the explicitly internal result in audits.yaml under AUDIT-RECORD.md;
never create a parallel audit document or claim independence. PASS supports otherwise-authorized operator validation,
not closure, merge or deployment.

From tigger-developer/sdlc, Apache-2.0.

More from tigger-developer/sdlc

Signals

Stars
0
Forks
0
Open issues
5
Last commit (this folder)
Oct 24, 2026 (5 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from tigger-developer

Cite this page

code-audit skill. Agent Mart. https://agentmart.dev/skills/tigger-developer-code-audit (last updated 2026-10-24)