integrate
Land an approved change: enforce the Done gate, fold in the specifications, archive the plan, merge, and close the loop in the tracker.
Skill by juangcarmona
Chalk Labs
Need help deploying integrate?
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/juangcarmona-integrate
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
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/integrate
What's inside
- Skill name
integrate- Description
- Land an approved change: enforce the Done gate, fold in the specifications, archive the plan, merge, and close the loop in the tracker. Use when integrating, landing or merging an approved work item.
- License
- MIT
- Metadata
- author: juangcarmona · plugin: agentic-sdlc
- Instructions
- 67 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
## Purpose Lifecycle stage 4. Turn an approved change into a **merged** one, with its specifications and its record landing alongside it. Input is a work item identifier, and it is required. If none was given, ask for one before starting. Invoke deliberately. No lifecycle stage runs this one on its own. ## When to use - Landing a change whose review has passed and whose Done gate is satisfied. - Closing a work item out in the tracker after its merge. - Publishing documentation the merged change made current. ## When not to use - Fixing what the Done gate reported unmet. That is `implement`. - Learning from the finished cycle. That is `review`, which this stage offers but never runs. - Rebasing a branch outside a merge. That is `rebase-safely` on its own. ## Workflow Every external write is shown first and confirmed. The merge most of all. An enabled `/ff` invocation may authorize routine lifecycle writes without per-write confirmation; it never waives required human approvals, branch protection, deterministic gates, or repository red lines. 1. **Load the configuration.** Read `docs/engineering-lifecycle.md`. Absent means stop and say to run `adopt` first. 2. **Refuse a change whose Done gate has not passed.** Delegate to the `done-gate-auditor` agent and require every applicable item satisfied, **both lanes**, with judgement items decided by their configured human roles or, in enabled `/ff` mode, by the independent reviewer only when explicitly marked substitutable. Unmet means stop, list what remains, and point at `implement`. The audit runs again here, in a fresh context, rather than trusting a verdict recorded earlier: this is the Definition of Done's enforcement point, and everything before it was preparation. 3. **Check the review.** `inspect-pull-request`: approved, no unresolved blocking threads, not a draft. In `/ff` mode an independent review decision may satisfy the lifecycle's substitutable review gate, but it never substitutes for approval required by the forge or repository policy. 4. **Bring the branch up to date with the target** via `rebase-safely`, so the specifications fold into current ones rather than stale ones. This is what keeps sequential merges from conflicting. 5. **Fold the specifications** via `sync-specs`, on the branch. Where the configuration's **Provided roles** names a skill for this contract, that skill is `sync-specs`; the contract is the same either way. 6. **Archive the plan artifact** via `archive-change`, on the branch, **after** the fold, never before. The same dispatch applies. 7. **Push** the branch, so the pull request carries the complete final state. 8. **Re-audit the Done gate on the final head.** Steps 4 through 7 rebased, folded, archived and pushed new commits, so the diff the step-2 audit covered is no longer the diff that will merge. Delegate to the `done-gate-auditor` agent again, in a fresh context, and require its verdict on the final head before proceeding. An audit of an earlier commit is not evidence for this one. 9. **Re-check the review immediately before merging.** Reviewer approval can be reset by a push, and steps 4 through 7 pushed. Read the pull request's state again and, in `/ff` mode, ensure the independent review covers the final head. A vote or model decision for an earlier commit is not evidence now. 10. **Merge** via `merge-pull-request`. Set the commit message **explicitly**: the subject describes the implemented change, not the proposal the pull request opened with, and no auto-generated body is allowed to carry anything unintended onto the target branch. 11. **Verify traceability.** The branch, the commits, the pull request and the artifacts all carry the work item identifier, and the item links to the merge. 12. **Close the loop in the tracker.** `transition-work-item` to the state the mapping names, and `comment-work-item` with the merge link. In `/ff` mode, include the run ID, completion status, merged commit, reviewer model and decisions, and the durable evidence path in that comment. 13. **Publish documentation** where a pair is repository-ahead, via `sync-doc-to-kb`. A pair reported diverged is refused: both sides changed, and either write destroys work. 14. **Offer the review now.** Draft the entry from what just happened, taking effort figures from `collect-usage`, and offer it while the human is still here. Declining is one word. This is the moment attention exists, and it never blocks the merge, which has already happened. Offering is not running: `review` runs when a human accepts. ## Validation - The Done gate was re-audited in this stage, not carried over from `implement`. - The Done gate was re-audited **again after the final push**, on the head that will merge, not on the pre-push commit. - The pull request was approved, not a draft, with no unresolved blocking threads, and its state was read again immediately before the merge. - The branch was brought up to date before the specifications were folded. - The plan was archived after the fold, never before. - The merge commit message was set explicitly. - The work item identifier appears on the branch, the commits, the pull request and the artifacts, and the item links to the merge. - No diverged documentation pair was written to. - The review was offered and not run. ## Common pitfalls | Pitfall | Fix | |---|---| | Merging on a Done gate verdict recorded earlier | Re-audit here. This is the enforcement point | | Merging on the pre-push audit after steps 4–7 pushed | Re-audit after the final push. The diff changed; the earlier verdict no longer covers it | | Trusting the review state over the gate | The gate is a hard precondition whatever the pull request says | | Merging a draft | Never | | Reading approval at the start and merging at the end | Steps 4 to 7 push, which can reset approval. Read it again | | Letting the forge auto-generate the merge body | Set it explicitly, describing the implemented change | | Folding specifications before rebasing | They fold into stale ones, and the next merge conflicts | | Archiving the plan before folding | Archive after, never before | | Writing to a diverged documentation pair | Refuse. Both sides changed and either write destroys work | | Running `review` because the merge succeeded | Offer it. A human accepts |
From juangcarmona/agent-toolkit, MIT.
More from juangcarmona/agent-toolkit
- 0.570score
adopt
juangcarmona
Bootstrap, reconcile, or explicitly reset a project's lifecycle configuration to the shared defaults.
SkillAFreshness: AClaude Code +9 - 0.570score
adr
juangcarmona
Draft and maintain Architecture Decision Records from maintainer-supplied decisions and repository evidence.
SkillAFreshness: AClaude Code +9 - 0.567score
architecture-docs
juangcarmona
Create, review, and maintain evidence-grounded arc42 architecture documentation from implementation evidence, ADRs, and product intent.
SkillAFreshness: AClaude Code +9 - 0.565score
review
juangcarmona
Learn from a merged cycle: record what happened with measured effort and route the lessons into the process itself.
SkillAFreshness: AClaude Code +9 - 0.565score
review-entry
juangcarmona
Create one dated, per-work-item retrospective after a change has merged: what worked, what did not, what to change next time, and sourced effort figures.
SkillAFreshness: AClaude Code +9 - 0.565score
status
juangcarmona
Report where a work item sits in the lifecycle, what comes next, and what has drifted.
SkillAFreshness: AClaude Code +9 - 0.565score
harden-process
juangcarmona
Turn one concrete failure into the smallest durable guard in the process itself.
SkillAFreshness: AClaude Code +9 - 0.565score
implement
juangcarmona
Apply the approved plan on its branch, verify the change against the Definition of Done in both lanes, and mark the pull request ready for review.
SkillAFreshness: AClaude Code +9 - 0.565score
propose
juangcarmona
Turn a ready work item into a reviewable solution design with tasks and test impact, raised as a draft pull request.
SkillAFreshness: AClaude Code +9 - 0.561score
ff
juangcarmona
Run the configured development lifecycle end to end with bounded, independent agent reviews and durable evidence.
SkillAFreshness: AClaude Code +9 - 0.553score
plan
juangcarmona
Plan a refined work item using the lifecycle's existing proposal workflow.
SkillAFreshness: AClaude Code +9 - 0.529score
document-to-markdown
juangcarmona
Convert Word (.doc, .docx), PowerPoint (.ppt, .pptx), Excel (.xls, .xlsx), OpenDocument (.odt, .ods, .odp), RTF, EPUB, CSV, and PDF files to GitHub-Flavored…
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 19, 2026 (11 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/juangcarmona/agent-toolkit
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from juangcarmona
- 0.570score
adr
juangcarmona
Draft and maintain Architecture Decision Records from maintainer-supplied decisions and repository evidence.
SkillAFreshness: AClaude Code +9 - 0.570score
adopt
juangcarmona
Bootstrap, reconcile, or explicitly reset a project's lifecycle configuration to the shared defaults.
SkillAFreshness: AClaude Code +9 - 0.567score
architecture-docs
juangcarmona
Create, review, and maintain evidence-grounded arc42 architecture documentation from implementation evidence, ADRs, and product intent.
SkillAFreshness: AClaude Code +9 - 0.565score
propose
juangcarmona
Turn a ready work item into a reviewable solution design with tasks and test impact, raised as a draft pull request.
SkillAFreshness: AClaude Code +9 - 0.565score
review
juangcarmona
Learn from a merged cycle: record what happened with measured effort and route the lessons into the process itself.
SkillAFreshness: AClaude Code +9 - 0.565score
harden-process
juangcarmona
Turn one concrete failure into the smallest durable guard in the process itself.
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 integrate 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
integrate skill. Agent Mart. https://agentmart.dev/skills/juangcarmona-integrate (last updated 2026-10-24)