review-taskgraph
Review taskgraph-related patches (transforms, kind YAML, worker config) for correctness, best practices, and reuse of existing upstream utilities
Skill by JohanLorenzo
Chalk Labs
Need help deploying review-taskgraph?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/johanlorenzo-review-taskgraph
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.
About
Reviews
Write a reviewNo 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
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/review-taskgraph
Tools it pre-approves 4
From allowed-tools in SKILL.md: the agent may use these without asking you each time.
Bash,Read,Grep,Glob
What's inside
- Skill name
review-taskgraph- Description
- Review taskgraph-related patches (transforms, kind YAML, worker config) for correctness, best practices, and reuse of existing upstream utilities
- License
- MPL-2.0
- Allowed tools
- Bash,, Read,, Grep,, Glob
- Instructions
- 91 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
- Pre-approves some shell commands
Files (4)
- SKILL.md
- checklist.md · references/
- reusable-transforms.md · references/
- get_diff.py · scripts/
SKILL.md (first lines)
# Review Taskgraph
Review taskgraph patches for correctness, best practices, and reuse of existing
upstream transforms and utilities.
## When to Use
Use this skill when the user invokes `/review-taskgraph` with one of:
- A GitHub PR URL (e.g., `https://github.com/org/repo/pull/42`)
- A Phabricator revision (e.g., `D12345` or full URL)
- A git commit range (e.g., `main..HEAD`)
- No argument (reviews uncommitted changes)
## Process
Note: if the user provided an argument, it is appended at the end of this skill
as `ARGUMENTS: <value>`. Use that value in step 1.
1. Get the diff. Check the `ARGUMENTS:` line at the bottom of this document.
- If `ARGUMENTS` is non-empty, run:
`python3 ~/.claude/skills/review-taskgraph/scripts/get_diff.py <ARGUMENTS_VALUE>`
- If `ARGUMENTS` is absent or empty, run:
`python3 ~/.claude/skills/review-taskgraph/scripts/get_diff.py`
2. From the diff, identify taskgraph-relevant files:
- Python files under `transforms/` directories
- Kind YAML files (`kind.yml` or `kind.yaml`)
- `taskcluster/config.yml`
- Dockerfiles under `taskcluster/docker/`
- Parameter files and `__init__.py` with `register()`
3. Read each modified file in full (not just diff hunks) to understand context.
4. Load the review references:
- @references/checklist.md
- @references/reusable-transforms.md
5. For each potential reuse opportunity, verify the utility actually exists by reading
the source at the paths from CLAUDE.local.md (`taskgraph_repo`,
`mozilla_taskgraph_repo`).
6. Produce the structured report below.
## Output Format
```
## Taskgraph Review: <target description>
### Verdict: PASS | PASS-WITH-NOTES | NEEDS-CHANGES
### Findings
#### `<file-path>`
- **[ERROR]** `<category>` (line N): <description>
- Fix: <concrete suggestion>
- **[WARNING]** `<category>` (line N): <description>
- Fix: <concrete suggestion>
- **[SUGGESTION]** `<category>` (line N): <description>
- Consider: <alternative approach>
### Reuse Opportunities
| Custom Code | Existing Utility | Import Path |
|---|---|---|
| <what the custom code does> | <utility name> | <import path> |
```
Severity levels:
- **ERROR**: will break at runtime, violates a hard constraint, or is a clear bug
- **WARNING**: works but violates conventions or misses a better approach
- **SUGGESTION**: style improvement or optional optimization
Verdict criteria:
- **NEEDS-CHANGES**: any ERROR or WARNING present
- **PASS-WITH-NOTES**: suggestions only, no errors or warnings
- **PASS**: no findings at all
Categories: `transform-correctness`, `kind-yaml`, `worker-scope`, `general-quality`,
`reuse`
## Important Notes
- This is read-only analysis. Do not modify any files.
- When uncertain about a finding, use SUGGESTION rather than ERROR.
- Skip files that are not taskgraph-related (e.g., application code, CI configs).
- If the diff is empty or contains no taskgraph files, report "No taskgraph changes found."From JohanLorenzo/claude-taskcluster-process, MPL-2.0.
More from JohanLorenzo/claude-taskcluster-process
- 0.565score
taskcluster-submit-task
JohanLorenzo
Directly submit a task to Taskcluster (Step 5 direct submission path).
SkillAFreshness: AClaude Code +9 - 0.565score
taskcluster-local-test
JohanLorenzo
Test a Taskcluster task locally before pushing to CI (Step 5 of the workflow).
SkillAFreshness: AClaude Code +9 - 0.561score
pr-description
JohanLorenzo
Write or revise a GitHub PR description or Phabricator test plan.
SkillAFreshness: AClaude Code +9 - 0.559score
taskcluster-monitor-group
JohanLorenzo
Monitor a Taskcluster CI group to completion and report failures.
SkillAFreshness: AClaude Code +9 - 0.558score
Git Origin Tracer
JohanLorenzo
Trace any line of code back to its original introduction, following moves, renames, and copies, then extract the bug number
SkillAFreshness: AClaude Code +9 - 0.558score
moz-phab
JohanLorenzo
Submit or update a Phabricator revision with moz-phab.
SkillAFreshness: AClaude Code +9 - 0.555score
Convert Github Permalink
JohanLorenzo
Convert local file references to GitHub permalinks using main/master branch SHA
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 30, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MPL-2.0
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from JohanLorenzo
- 0.565score
taskcluster-submit-task
JohanLorenzo
Directly submit a task to Taskcluster (Step 5 direct submission path).
SkillAFreshness: AClaude Code +9 - 0.565score
taskcluster-local-test
JohanLorenzo
Test a Taskcluster task locally before pushing to CI (Step 5 of the workflow).
SkillAFreshness: AClaude Code +9 - 0.561score
pr-description
JohanLorenzo
Write or revise a GitHub PR description or Phabricator test plan.
SkillAFreshness: AClaude Code +9 - 0.559score
taskcluster-monitor-group
JohanLorenzo
Monitor a Taskcluster CI group to completion and report failures.
SkillAFreshness: AClaude Code +9 - 0.558score
Git Origin Tracer
JohanLorenzo
Trace any line of code back to its original introduction, following moves, renames, and copies, then extract the bug number
SkillAFreshness: AClaude Code +9 - 0.558score
moz-phab
JohanLorenzo
Submit or update a Phabricator revision with moz-phab.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get review-taskgraph running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
review-taskgraph skill. Agent Mart. https://agentmart.dev/skills/johanlorenzo-review-taskgraph (last updated 2026-10-24)