ftk-execute
Execute a Spec's Work Items subagent-driven — a fresh ftk-implementer per item, ftk-task-reviewer gate per item, final whole-run review.
Skill by adiwirya
Chalk Labs
Need help deploying ftk-execute?
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/adiwirya-ftk-execute
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
Use cases: productivity, agent-building
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/ftk-execute
What's inside
- Skill name
ftk-execute- Description
- Execute a Spec's Work Items subagent-driven — a fresh ftk-implementer per item, ftk-task-reviewer gate per item, final whole-run review. Use when a Spec has multiple approved Work Items to implement.
- License
- MIT
- Instructions
- 40 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
Orchestrate Work Item execution with fresh-context subagents. You coordinate only — never implement or fix anything yourself; all code changes go through dispatched agents. If this environment has no subagent dispatch tool, fall back: execute items sequentially in this session following `ftk-implement` for each, with a manual review pass between items. ## 1. Preflight - `.ftk/config.yaml` must be valid — if missing/invalid, tell the user to run `ftk-config` first and stop. - `git status --porcelain` must be clean — if not, stop and ask. - On the repo's default branch? Create a work branch first (`feat/<spec-slug>`). - Ask the user whether commits are allowed this run; pass that decision verbatim into every implementer dispatch. If commits are forbidden, per-item diffs use tree snapshots instead of commits (step 4.1) and warn the user that resume relies on the ledger's recorded hashes, not git log. ## 2. Load and order Read the Spec and all its Work Items. Order topologically by `## Blocked by`. If any item has unanswered `## Blocking decisions`, stop and ask those questions before dispatching anything. ## 3. Ledger Create or read `.ftk/execute-progress.md`. Items already recorded as complete are DONE — never re-dispatch them; resume at the first unrecorded item. After a session break, trust the ledger and `git log` over conversation memory. ## 4. Per item, sequentially 1. Record the base snapshot: with commits allowed, `git rev-parse HEAD`; with commits forbidden, `git add -A && git write-tree` (a tree hash — creates no commit). 2. Dispatch **ftk-implementer** with: the full work item body, the project root, the commits-allowed decision, and interfaces/decisions from earlier items it cannot know. Do not paste prior items' full history. 3. Handle its status: - `DONE` → review. `DONE_WITH_CONCERNS` → read the concerns first; correctness/scope concerns get resolved before review. - `NEEDS_CONTEXT` → supply the missing context, re-dispatch. - `BLOCKED` → change something (context, decomposition, or escalate to the user). Never re-dispatch unchanged. 4. Dispatch **ftk-task-reviewer** with the work item body, the diff range (`<base>..HEAD` when committing; in no-commit mode take a head snapshot the same way as step 1 and pass both tree hashes for `git diff <baseTree> <headTree>`), and the implementer report. 5. Gate: Critical/Important findings or a FAIL spec verdict → dispatch a fix to ftk-implementer (same item, findings list, covering tests named), then re-review. Repeat until both verdicts pass. Record Minor findings in the ledger for final-review triage. 6. Mark the item's acceptance criteria checked in its file (local storage) and append the ledger line: `<item>: complete (<base7>..<head7>, review clean — Minors deferred)` (hashes are commits, or tree snapshots in no-commit mode). ## 5. Final review After all items: dispatch **ftk-task-reviewer** in final-review mode with the Spec, the full run diff range, and the ledger's Minor list. `NOT READY` findings go back through the fix loop. Stop only at `READY`. ## 6. Report Table: work item → commit range → verdicts. List deferred Minors. Recommend `ftk-commit` / `ftk-pr` as next steps — do not invoke them automatically.
From adiwirya/flutter-tools-kit, MIT.
More from adiwirya/flutter-tools-kit
- 0.479score
ftk-interview
adiwirya
Resolve intent, language, constraints, and decision dependencies before writing a Spec.
SkillCFreshness: CClaude Code +9 - 0.479score
ftk-create-spec
adiwirya
Create and save a Spec from current conversation context.
SkillCFreshness: CClaude Code +9 - 0.479score
ftk-config
adiwirya
Configure FTK workflow storage and semantics for the current working directory.
SkillCFreshness: CClaude Code +9 - 0.476score
ftk-build-ui
adiwirya
Build a Flutter screen or widget that visually matches a design from Figma, an image, or a description.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-ci
adiwirya
Generate a GitHub Actions workflow for a Flutter project — analyze, test, smoke build, artifact.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-scaffold-feature
adiwirya
Add a feature module (data/domain/presentation + Riverpod providers + route) to an existing Flutter project.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-scaffold-project
adiwirya
Create a new Flutter project with feature-first structure, Riverpod, linting, and optional flavors.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-l10n
adiwirya
Set up Flutter localization (ARB + intl, Indonesian/English default) or migrate hardcoded strings to AppLocalizations.
SkillCFreshness: CClaude Code +9 - 0.471score
ftk-refactor-legacy
adiwirya
Migrate an existing Flutter project toward FTK standards (feature-first, Riverpod, dio) one feature at a time, behind characterization tests.
SkillCFreshness: CClaude Code +9 - 0.469score
ftk-release
adiwirya
Build a release of a Flutter app — version bump, signing checks, APK/AAB/IPA build, upload guidance.
SkillCFreshness: CClaude Code +9 - 0.469score
ftk-add-tests
adiwirya
Add unit, widget, golden, or integration tests to a Flutter project, choosing the right level per seam.
SkillCFreshness: CClaude Code +9 - 0.469score
ftk-perf
adiwirya
Diagnose and fix Flutter performance problems — jank, slow lists, slow startup — measurement-first in profile mode.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 10, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/adiwirya/flutter-tools-kit
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from adiwirya
- 0.479score
ftk-create-spec
adiwirya
Create and save a Spec from current conversation context.
SkillCFreshness: CClaude Code +9 - 0.479score
ftk-config
adiwirya
Configure FTK workflow storage and semantics for the current working directory.
SkillCFreshness: CClaude Code +9 - 0.479score
ftk-interview
adiwirya
Resolve intent, language, constraints, and decision dependencies before writing a Spec.
SkillCFreshness: CClaude Code +9 - 0.476score
ftk-build-ui
adiwirya
Build a Flutter screen or widget that visually matches a design from Figma, an image, or a description.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-ci
adiwirya
Generate a GitHub Actions workflow for a Flutter project — analyze, test, smoke build, artifact.
SkillCFreshness: CClaude Code +9 - 0.474score
ftk-l10n
adiwirya
Set up Flutter localization (ARB + intl, Indonesian/English default) or migrate hardcoded strings to AppLocalizations.
SkillCFreshness: CClaude 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 ftk-execute 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
ftk-execute skill. Agent Mart. https://agentmart.dev/skills/adiwirya-ftk-execute (last updated 2026-10-24)