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/harshanandak/forge/protected-commitnpx skills add harshanandak/forge --skill protected-commitgit clone --depth 1 https://github.com/harshanandak/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 | $0.00082 | $0.00991 |
| Opus 5 | $0.00041 | $0.00495 |
| Sonnet 5 | $0.00016 | $0.00198 |
| Haiku 4.5 | $0.00008 | $0.00099 |
Grade A, and why
protected-commit 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
protected-commit
The gate is content-bound and fails closed. Read what it told you, then use the owning command — or stop and say the path is currently unreachable.
What the gate actually is
scripts/protected-state-check.js runs on every pre-commit (lefthook job
protected-state, no glob, so it sees every commit). For each staged file it
calls assertProtectedWriteAllowed, and any file that matches a protected
surface needs a one-time, content-bound Kernel capability issued for the
exact actor, worktree, surface, path, bytes, and source HEAD.
[verified 2026-08-13 — lib/protected-state-authority.js]
Ask the code which surface a path is on rather than guessing:
node -e "console.log(require('./lib/protected-state-surfaces').assertProtectedWriteAllowed('AGENTS.md'))"
Three things that are commonly believed and are wrong
-
FORGE_PROTECTED_STATE_ALLOWED_SURFACESdoes not authorize anything. No runtime code path reads it. It appears only in prose docs and one test fixture.docs/reference/protected-state-surfaces.mdsays so itself at the "Surface-only environment declarations do not authorize protected changes" line, and then contradicts itself further down — the contradiction is a documentation bug, not a hidden door. [verified 2026-08-13] -
The
categories:block in.forge/protected-paths.yamlis not what runs. Runtime enforcement uses a hardcodedPROTECTED_SURFACESlist inlib/protected-state-surfaces.js. The manifest's W1 categories (forge_core,user_protocol,generated_artifacts) are documentation. Solib/**,scripts/**, andskills/**— listed underforge_corein the manifest — are not blocked at runtime today, whileAGENTS.mdis, under the legacy surface idgenerated_harness. [verified 2026-08-13] -
--no-verifyis not the answer. Bypassing hooks is forbidden for agents in this repo, and a hook you route around protects nobody. If the gate is wrong, the gate gets fixed in its own PR. [Forge #9 ×3]
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 · 82 lines · 82 tokens per session scan A d8ed668a8250
protected-commit is a skill published in the GitHub repository harshanandak/forge (4 stars, last pushed 2d ago), licensed MIT. It adds 82 tokens to every session and 991 once invoked, about $0.0004 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…