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 agentmods add skills/dormstern/forge/stabilizationnpx skills add dormstern/forge --skill stabilizationgit clone --depth 1 https://github.com/dormstern/forgeWhat 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 | $0.00098 | $0.01196 |
| Opus 5 | $0.00049 | $0.00598 |
| Sonnet 5 | $0.00020 | $0.00239 |
| Haiku 4.5 | $0.00010 | $0.00120 |
Grade A, and why
stabilization 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 yesterday.
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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stabilization Triggers
Stabilization is the framework's circuit breaker. When build quality slips below the threshold, you stop adding new features and pay down the rework + debt that's accumulated. One bad release is fine. Two without intervention is a pattern. Three without stabilization is a maintenance crisis.
The four triggers
Each is computed at SHAPE time across the last release (or last 5 features if mid-cycle):
| Trigger | Threshold | Why it matters |
|---|---|---|
| Rework rate | > 15% | Code shipped recently is being modified — the spec was wrong |
| Blocked features | ≥ 3 | Architecture is unclear or features are oversized |
| Avg builder resets | > 2.0 per feature | Tests are too tight or domain is unclear |
| Avg quality_score | < 6.0 across last 5 | Systemic quality degradation |
If ANY trigger fires, the orchestrator recommends a stabilization release in SHAPE.
What a stabilization release looks like
- Thesis: "Pay down the rework + debt accumulated in v{X}."
- Features: 0 new
- Debt items: every TECH_DEBT flag from the last 2 releases
- Rework: every REWORK flag, with proper fixes
- Removals: anything from the kill list that hasn't shipped
- Identity: no changes
- Drift: full audit at the end
Stabilization releases are short — typically half the duration of a normal release. They're not glamorous. Nobody tweets about a stabilization release. They're load-bearing.
Why each trigger matters
Rework rate > 15%
If 3 of 20 features in the last release modified code that just shipped, you're shipping the wrong thing. Possible causes:
- Specs were unclear → test-writer wrote tests that allowed wrong behavior to pass
- Architecture was misunderstood → builder placed code in the wrong module
- IDENTITY drifted → the feature didn't actually serve WHO/JOB
- Contracts were violated → the breaking change is rippling
Stabilization gives you space to find the upstream cause without shipping more rework.
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.
- yesterday First seen · 110 lines · 98 tokens per session scan A 432ff6865a30
stabilization is a skill published in the GitHub repository dormstern/forge (6 stars, last pushed 3mo ago), licensed MIT. It adds 98 tokens to every session and 1,196 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…