Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add marcusgoll/Spec-Flow --skill git-workflow-enforcergit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/marcusgoll/spec-flow/git-workflow-enforcer)<a href="https://agentmods.dev/skills/marcusgoll/spec-flow/git-workflow-enforcer"><img src="https://agentmods.dev/badge/skills/marcusgoll/spec-flow/git-workflow-enforcer.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00069 | $0.03334 |
| Opus 5 | $0.00034 | $0.01667 |
| Sonnet 5 | $0.00014 | $0.00667 |
| Haiku 4.5 | $0.00007 | $0.00333 |
Grade A, and why
git-workflow-enforcer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 442 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<quick_start> <auto_commit_workflow> Automatic commit generation when changes detected:
- Detect uncommitted changes: Run
git status --porcelain - Classify change type: Phase completion, task completion, or file modification
- Generate commit message: Use Conventional Commits format based on context
- Validate safety: Check branch, merge conflicts, commit message format
- Execute commit: Stage files and commit with generated message
- Provide feedback: Show commit hash and rollback command
Example:
Context: Task T001 completed
❌ UNCOMMITTED CHANGES DETECTED
Modified: api/models/message.py
Created: api/tests/test_message.py
✅ AUTO-COMMITTING with generated message:
feat(green): T001 implement Message model to pass test
Implementation: SQLAlchemy model with validation
Tests: 26/26 passing
Coverage: 93% (+1%)
✅ COMMITTED: abc123f
Rollback: git revert abc123f
</auto_commit_workflow>
<trigger_patterns> Auto-invoke when detecting:
Phase completion markers:
- "Phase N complete"
- "/specify complete", "/plan complete", "/tasks complete"
- "analysis complete", "optimization complete"
- Phase command finishing
Task completion markers:
- "T### complete", "task ### done"
- "task-tracker mark-done" command
- "mark task complete"
Change detection:
git status --porcelainreturns non-empty- Files created/modified/deleted
- Working tree dirty before phase transition </trigger_patterns>
<commit_message_generation> Conventional Commits format:
<type>(<scope>): <subject>
[optional body]
[optional footer]
Auto-detection logic:
| Context | Type | Scope | Subject Template |
|---|---|---|---|
| Phase 0 | docs | spec | create specification for {feature} |
| Phase 0.5 | docs | clarify | resolve {n} clarifications for {feature} |
| Phase 1 | docs | plan | create implementation plan for {feature} |
| Phase 2 | docs | tasks | create task breakdown for {feature} ({n} tasks) |
| Phase 3 | docs | analyze | create cross-artifact analysis for {feature} |
| Phase 4 (RED) | test | red | {taskId} write failing test for {description} |
| Phase 4 (GREEN) | feat | green | {taskId} implement {description} to pass test |
| Phase 4 (REFACTOR) | refactor | - | {taskId} improve {description} |
| Phase 5 | docs | optimize | complete optimization review for {feature} |
| Phase 6 | docs | preview | create release notes for {feature} v{version} |
Body generation:
- Phase commits: Include metrics (task count, criteria count, etc.)
- Task commits: Include evidence (tests, coverage, commit hash)
- Fix commits: Include root cause and solution
See references/commit-templates.md for complete templates. </commit_message_generation> </quick_start>
Check for uncommitted changes when:
- Phase command completes
- task-tracker mark-done-with-notes called
- User mentions "commit", "save progress", etc.
# Check for uncommitted changes
git status --porcelain
# If non-empty output → uncommitted changes exist
</detection_phase>
<classification_phase> 2. Classify Change Type and Extract Context
Phase completion:
- Extract phase name from workflow state or command
- Extract feature slug from current directory
- Get artifact file paths (spec.md, plan.md, etc.)
Task completion:
- Extract TaskId from task-tracker parameters or conversation
- Extract task description from tasks.md
- Get phase marker ([RED], [GREEN], [REFACTOR])
- Extract evidence (tests, coverage) from completion context
File modification:
- Analyze changed files to infer purpose
- Check git log for recent commit patterns
- Default to "chore" type if unclear </classification_phase>
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 442 lines · 69 tokens per session scan A af2aa0fa47d4
git-workflow-enforcer is a skill published in the GitHub repository marcusgoll/Spec-Flow (92 stars, last pushed 4mo ago), licensed MIT. It adds 69 tokens to every session and 3,334 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
skill-security-hooks
Install and maintain defensive security hooks and CI gates for authorized SaaS projects. Use when asked to add pre-commit hooks, pre-push hooks, CI security scans, GitHub Actions hardening, secret scanning, dependency scanning, SAST, DAST gates, or security automation around existing repositories.
checkin-assistant
A checklist for the steps before making a Git commit. It treats a commit as one complete, reversible unit of work that should leave the codebase working.
commit-standards
Format commit messages following conventional commits standard. Use when: writing commit messages, git commit, reviewing commit history. Keywords: commit, git, message, conventional, feat, fix, refactor.
setup-pre-commit
Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo. Use when user wants to add pre-commit hooks, set up Husky, configure lint-staged, or add commit-time formatting/typechecking/testing.
atmos-git
Atmos Git and GitOps: git.repositories, clone/pull/status/diff/commit/push/clean, local Git hook shims, signed commits, managed workdirs, fork-PR trust gate, and auth via identities or github/sts.
security-pipeline
Use when security verification is needed - pre-commit security checks, vulnerability scanning, STRIDE threat analysis. Integrates with /handoff-verify --security and /commit-push-pr. CWE Top 25 based.