artifact-commit-verification
Class-level artifact and commit verification for claimed agent work, exact file diffs, and closeout evidence.
Best use case
artifact-commit-verification is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Class-level artifact and commit verification for claimed agent work, exact file diffs, and closeout evidence.
Teams using artifact-commit-verification should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/artifact-commit-verification/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How artifact-commit-verification Compares
| Feature / Agent | artifact-commit-verification | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Class-level artifact and commit verification for claimed agent work, exact file diffs, and closeout evidence.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Artifact Commit Verification ## When to Use Use when an agent claims work is complete and you need to verify exact files, commits, generated artifacts, issue closeout evidence, or working-tree state before trusting the claim. ## Class-Level Workflow 1. Identify the claimed artifact/file/commit set. 2. Compare the exact claimed paths against git status, git diff, and generated output files. 3. Confirm no unrelated dirt is being mistaken for task output. 4. For public/distributable generated artifacts, inspect machine-readable sidecars/manifests as well as the visible output: they must not serialize absolute local paths, staging roots, client repo names, temp directories, or other environment/client identifiers. 5. For visual artifacts (PNG/SVG/PDF/HTML charts, brochures, dashboards), run both structural smoke checks and a human/visual readability check before claiming success. 6. If the repo-wide/legal scanner fails because of unrelated dirty files outside the claimed artifact set, do not either ignore it or falsely report a clean official scan. Run a staged/claimed-file-only deny-list check, report it explicitly as scoped evidence, and document the unrelated blocker separately. 7. Record evidence before closing or reporting success. ## Public Artifact Hygiene When a task generates client-facing or public GTM/report assets, treat metadata files as part of the artifact surface. Search generated manifests/sidecars and representative rendered files for `/mnt/`, `/tmp/`, workspace names, confidential repo buckets, and client/project identifiers that are not intended for publication. If an adversarial review flags local path leakage, fix the generator so future outputs emit repo-relative or artifact-relative paths rather than post-processing one file by hand. See `references/public-chart-artifact-hygiene.md` for the #2555 chart-pack example. When the official scanner reports failures from unrelated uncommitted state (for example `.Codex/state/**` session logs), preserve truthfulness by separating: - **official scan result:** failed/blocked and why - **scoped staged-file evidence:** exact files scanned and pass/fail result - **decision:** whether the scoped pass is sufficient for a docs-only/handoff commit, or whether the unrelated blocker must be cleaned before publication/release Never summarize a scoped staged-only scan as a full repo legal PASS. ## Consolidated Session Learnings Narrow skills absorbed during the 2026-04-29 umbrella consolidation are preserved under `references/`. ## Absorbed Narrow Skills (2026-04-29) ### `targeted-artifact-commit-verification` - Former skill demoted to `references/targeted-artifact-commit-verification.md`. - Preserved insight: Verify whether the exact files from a just-completed task are still uncommitted before creating another commit, especially in dirty repos with unrelated churn.
Related Skills
subagent-write-verification
Independently verify subagent-claimed file writes with filesystem and git checks before treating the artifact as real, before committing it, and before referencing the path in downstream prompts.
verify-Codex-run-commit-vs-working-tree-before-closing
After a Codex implementation run, verify the claimed file set against the actual commit and working tree before treating the issue as fully complete.
tdd-verification-and-adversarial-review
Verify pre-written TDD tests pass, conduct adversarial code review on committed diffs, and route findings to existing issues
tax-form-navigation-verification
Systematic workflow for verifying tax software calculations against entry guide specifications
portable-pattern-verification-workflow
Multi-package implementation with verification strategy for cross-platform configuration hardening
multi-year-tax-filing-verification-workflow
Verify and reconcile complex multi-form tax filings by cross-referencing source documents, identifying data dependencies, and validating line-by-line against authoritative records.
evaluate-local-commits-via-cherry-pick-dry-run
Technique to identify which ahead commits contain real changes vs. already-merged or ephemeral content
adversarial-code-review-for-committed-diffs
Systematic process for reviewing already-committed code changes to catch type inconsistencies, edge cases, and docstring gaps
static-site-build-artifact-plan-review
Plan-review pattern for static-site fixes where the deployed artifact is generated from source files (e.g. sitemap/robots/static assets). Prevents review churn by separating durable regression checks from one-time migration verification and by validating built output, not just source files.
plan-review-handoff-verification
Verify a parked GitHub issue plan-review handoff across live labels, local markers, README rows, canonical review artifacts, and stale plan-body evidence before reporting next action.
plan-review-artifact-authority-and-approval-drift
Keep iterative plan-review artifacts truthful when external reruns overtake self-reviews or stale approval signals remain on older revisions.
plan-draft-review-artifact-truthfulness-and-issue-body-alignment
Keep plan drafts truthful during adversarial review loops by verifying real provider artifact state on disk and aligning the GitHub issue body to the bounded plan tranche before claiming approval-readiness.